In recent years, adversarial attacks in hyperspectral image (HSI) classification have garnered increasing attention. However, existing attack methods primarily manipulate individual pixel spectral to mislead deep neural networks (DNNs) into misclassification, overlooking the physical consistency of hyperspectral data. This oversight results in adversarial samples that lack physical interpretability and suffer from low attack efficiency. To alleviate these issues, this paper proposes a sparse unmixing guided adversarial attack framework (SUGAA) to efficiently generate hyperspectral adversarial samples that satisfy physical consistency. The proposed framework first employs sparse unmixing to extract the abundance matrix of HSI, introducing adversarial perturbations to the abundance matrix to generate physically consistent adversarial samples. Additionally, SUGAA leverages the compositional similarity of materials within intra-class HSI pixels to design a class-specific perturbation generation strategy, enhancing the applicability of adversarial perturbations across pixels of the same class. To further improve optimization effectiveness, SUGAA incorporates a class-specific perturbation optimization algorithm based on momentum iterative gradients to avoid local optima, ensuring stable and efficient perturbation generation. Experimental results on real HSI datasets demonstrate that SUGAA not only generates adversarial samples with high attack performance and physical consistency but also exhibits robustness to common preprocessing transformations.

Large language models (LLMs) have demonstrated remarkable capabilities across various natural language processing tasks, but they remain vulnerable to adversarial attacks and pose significant security concerns. Existing attack methods often treat adversarial prompts as flat sequences, neglecting the rich hierarchical structure of natural language, which could limit their effectiveness. Advancing the methodologies for adversarial attacks is crucial for rigorously assessing the security of LLMs and identifying subtle vulnerabilities. This paper introduces AdvGP, a novel framework that leverages genetic programming (GP) to generate adversarial prompts for LLMs. AdvGP exploits the inherent structural similarities between GP trees and natural language syntax to optimize the structure of harmful prompts. The framework incorporates a multi-granularity hierarchical attack strategy, specialized genetic operators that leverage an assisting LLM for depth-aware crossover and multi-level mutation, and a comprehensive fitness function integrating semantic consistency and attack effectiveness. The proposed method achieves competitive attack performance on multiple LLMs, consistently generating harmful outputs despite higher perplexity than some baselines. Ablation studies confirm the significant contributions of both LLM-aided and depth-aware mechanisms to AdvGP’s effectiveness. Furthermore, transferability analysis reveals that the generated prompts are able to bypass the defenses of various state-of-the-art LLMs, such as ChatGPT and Gemini.

For expensive multiobjective optimization problems, there exists useful knowledge, e.g., the trained surrogate models, can be transferred to assist the optimization of a target optimization problem, which is termed as multi-problem surrogates. Stacking transfer is able to combine the pretrained source surrogate models and the preliminary target model with a meta-regression algorithm to transfer knowledge from source to target. However, when large-scale source models are involved in the many-problem scenarios, the less correlated sources may hurt the target performance, which is known as negative transfer. In this paper, sparse representation of the coefficients of meta-regression is considered to automatically select the most relevant source models for largely avoiding negative transfer. In the proposed many-problem surrogates, the coefficients of the source and target models are assumed to be sparse under the non-negativity and sum-to-one constraints. Then a sparse transfer stacking model is established with l1-norm of the coefficients. Next, the alternating direction method of multipliers is employed to solve the resulting constrained optimization problem by converting it into several much simpler problems. Most of the previous works assume that the costs for evaluation have no much difference and this assumption rarely holds in the real-world applications. In order to further reduce the total costs, an improved surrogate model with a cost-sensitive measure is designed to estimate the cost and select new solutions for real evaluation based on their estimated fitness, uncertainty and cost. Experimental results on synthetic and practical problems have demonstrated the superiority of the proposed many-problem surrogates.

Data-driven evolutionary algorithms (DDEAs) have achieved significant success in numerous real-world optimization problems, where exact objective functions and constraint functions do not exist, and they mainly rely on available data. However, the existing DDEAs primarily focus on improving performance through data and surrogate, without considering that the users may lack the specialized domain knowledge and sufficient computing resources required for DDEAs. To address the aforementioned issues, this paper proposes a novel paradigm called Evolutionary Learning and Optimization as a Service (ELOaaS) and investigates the potential collusion attacks between machine learning modules and evolutionary computing modules on cloud server, which may lead to privacy leakage. Consequently, a privacy-enhanced DDEA (PEDDEA) is proposed as an instantiation algorithm of ELOaaS, which is designed to tackle offline data-driven evolutionary optimization within the ELOaaS paradigm. In the proposed PEDDEA, a subspace learning-based privacy protection strategy is designed to defense the collusion attacks. Additionally, a model management strategy based on Kendall tau metric is introduced to construct high-quality surrogate ensembles. PEDDEA enables users to outsource private offline data to cloud servers, thereby approaching the optimal solution while ensuring privacy protection. Comprehensive experiments are conducted on benchmark problems and safety evaluation problems of autonomous vehicles. According to the experimental results, the proposed algorithm has significant performance advantages over existing offline DDEAs while ensuring privacy protection.

Federated Learning (FL) has emerged as a privacy-preserving paradigm for training machine learning models across distributed edge devices in the Internet of Things (IoT). By keeping data local and coordinating model training through a central server, FL effectively addresses privacy concerns and reduces communication overhead. However, the limited computational power, memory, and bandwidth of IoT edge devices pose significant challenges to the efficiency and scalability of FL, especially when training deep neural networks. Various FL frameworks have been proposed to reduce computation and communication overheads through dropout or layer freezing. However, these approaches often sacrifice accuracy or neglect memory constraints. To this end, in this work, we introduce Federated Learning with Ordered Layer Freezing (FedOLF). FedOLF consistently freezes layers in a predefined order before training, significantly mitigating computation and memory requirements. To further reduce communication and energy costs, we incorporate Tensor Operation Approximation (TOA), a lightweight alternative to conventional quantization that better preserves model accuracy. Experimental results demonstrate that over non-iid data, FedOLF achieves at least 0.3%, 6.4%, 5.81%, 4.4%, 6.27% and 1.29% higher accuracy than existing works respectively on EMNIST (with CNN), CIFAR-10 (with AlexNet), CIFAR-100 (with ResNet20 and ResNet44), and CINIC-10 (with ResNet20 and ResNet44), along with higher energy efficiency and lower memory footprint.

Personalized Federated Learning (PFL) is widely employed in the Internet of Things (IoT) to handle high-volume, non-iid client data while ensuring data privacy. However, heterogeneous edge devices owned by clients may impose varying degrees of resource constraints, causing computation and communication bottlenecks for PFL. Federated Dropout has emerged as a popular strategy to address this challenge, wherein only a subset of the global model, i.e. a sub-model, is trained on a client's device, thereby reducing computation and communication overheads. Nevertheless, the dropout-based model-pruning strategy may introduce bias, particularly towards non-iid local data. When biased sub-models absorb highly divergent parameters from other clients, performance degradation becomes inevitable. In response, we propose federated learning with stochastic parameter update (FedSPU). Unlike dropout that tailors local models to small-size sub-models, FedSPU maintains the full model architecture on each device but randomly freezes a certain percentage of neurons in the local model during training while updating the remaining neurons. This approach ensures that a portion of the local model remains personalized, thereby enhancing the model's robustness against biased parameters from other clients. Experimental results demonstrate that FedSPU outperforms federated dropout by 4.45% on average in terms of accuracy. Furthermore, an introduced early stopping scheme leads to a significant reduction of the training time in FedSPU by 25%~71% while maintaining high accuracy.

Wind farms, designed based on certain assumed wind speed and direction patterns, may experience sub-optimal power generation in changing weather conditions due to turbine yaw misalignment. Accordingly, optimal control of turbine yaw angles to maximize overall power generation across all turbines in the farm, while mitigating wake effects caused by upstream turbines, becomes essential. It is typically formulated as a model predictive control (MPC) problem. A key module in this MPC problem is rolling horizon optimization, which is often nonlinear and non-convex, posing challenges to traditional mathematical programming solvers. This work reformulates rolling horizon optimization into a multi-task optimization (MTO) problem, where each component task is defined as maximizing the power generation of an individual turbine while accounting for wake effects from its upstream turbines. All component tasks are co-solved, with inter-task knowledge transfer and reuse, to enhance total power generation. We customize an existing evolutionary MTO method to solve the problem, using yaw angle adjustments as the means of knowledge transfer. The proposed approach is tested on the Penmanshiel wind farm layout with 14 turbines, outperforming the commonly used traditional solver.

The cold-start problem remains a long-standing challenge in recommender systems. Recent advances in large language models (LLMs) have opened new avenues for addressing cold-start scenarios through data augmentation. However, existing cold-start augmentation methods often suffer from negative augmentation, manifesting as incomplete augmentation, where generated interactions fail to comprehensively reflect user preferences, and inaccurate augmentation, where they conflict with user intent. These issues largely stem from two limitations: (1) the inability to effectively incorporate collaborative signals, which are critical for preference alignment, and (2) the lack of awareness of the downstream model's learning dynamics during data augmentation. To the best of our knowledge, the latter has not been studied in the literature.

Research on adversarial attacks in remote sensing tasks have predominantly focused on designing perturbations or patches, presenting challenges in balancing attack success rate and adversarial stealthiness. Instead of focusing on designing adversarial examples under adversarial perturbation constraints to ensure stealthiness, this paper proposes a Style Transfer Enabled Adversarial Attack with Attention Mechanism (STEAM), which leverages style transfer to generate adversarial examples with high visual fidelity. Specifically, STEAM transfers distinctive styles from critical regions of source samples to attackable areas in target samples, effectively incorporating natural textures from source samples. To further refine this process, an attention mechanism is introduced to selectively extract style features from key regions of the source samples, mitigating redundancy from global style information. Additionally, selective style transfer process also includes the consideration of semantic features across different regions in target samples, ensuring a more effective attack area selection. As a result, STEAM achieves high attack success rate by utilizing proper style selected from groups of style samples, and preserving high visual fidelity through selectively transfer the natural style feature into specific attackable region in target samples. Experimental results on the UCM and WHU-RS19 datasets demonstrate that STEAM not only enhances the visual fidelity of adversarial examples but also improves the attack success rate. Furthermore, experiments against state-of-the-art adversarial defense methods highlight the adversarial attack effectiveness and robostness of STEAM compared to other adversarial attack methods.

Graph matching aims to establish node correspondences between graphs, which is a classic combinatorial optimization problem. In recent years, (deep) learning-based methods have emerged as a superior alternative to traditional graph matching solvers. However, these methods typically rely on node-level correspondence labels, which can be prohibitively expensive or unrealistic. Inspired by contrastive learning that is a prevalent paradigm for self-supervised representation learning, we develop a Contrastive Learning Network for Unsupervised Graph Matching (CUGM), which is an end-to-end differentiable pipeline to learn node permutations. Specifically, we propose three-level augmentation including raw image augmentation, graph augmentation and model augmentation for generating diverse enough contrastive views to enrich training instances. Then a contrastive learning network is constructed to capture the higher-order structural information in graphs and learn the final node representations for yielding the affinity matrix to directly solve a linear assignment problem. More importantly, we propose a node-level contrastive loss with false negative cancellation for optimizing the whole network to extract the tailored node feature representations to improve graph matching accuracy. Experimental results on standard graph matching benchmarks demonstrate that our end-to-end unsupervised method achieves the competitive performance compared with state-of-the-art supervised and unsupervised graph matching methods.

Transfer evolutionary multiobjective optimization leverages the relevant knowledge from other source problems (distinct but possibly related) to assist the optimization of the target problem of interest. Multi-problem surrogates stack multiple source surrogates to reduce the number of function evaluations of the target expensive problem. The current multi-problem surrogates only considers several source problems and the source and target problems are assumed to be homogeneous. In order to address the above issues, this paper proposes fast heterogeneous multi-problem surrogates for transfer evolutionary multiobjective optimization with a large number of surrogates. First, an iterative surrogate selection strategy is designed to select the highly relevant surrogates from the large-scale surrogate pool to avoid negative transfer. Second, heterogeneous multi-problem surrogates are established to align the features of the source and target models. Finally, an adaptive k-fold cross-validation method is proposed to obtain the predicted values of the target model with low computational costs. Experiments on the multiobjective optimization benchmark problems and multiobjective neural architecture search problems have demonstrated that the proposed method is able to avoid negative transfer in the large-scale scenarios and reduce the computational costs.

For adversarial attacks on aerial image object detection, some physical background attack methods have been proposed and demonstrated excellent performance. However, most of the methods restrict the protected object to a fixed area, and changes in the position and size of the object can affect the effectiveness of the attack. In addition, the size of the background is only related to the size of the object, without considering any reduction in the background area. To alleviate these issues, this paper proposes a novel adversarial background patch generation strategy, which generates an adversarial background patch where aircrafts parked at any position within it remain undetected. Besides we also aim to make the size of the background as small as possible to improve the concealment of the attack and reduce the economic cost. Specifically, a series of adversarial images are generated by placing the aircraft at random angles and positions on the adversarial background patch. Then, a novel background patch optimization strategy is proposed, which enhances the occlusion robustness of the adversarial background patch by weighting the adversarial loss of each image in the batch based on adversarial difficulty. In addition, this paper proposes a novel area loss for achieving optimal area size, which is used to reduce the cost of producing the patch. Finally, a multi-objective optimization method based on Pareto efficiency is introduced for balancing two conflicting losses, the adversarial loss and the area loss. Experimental results show that the adversarial background patches generated by this method have excellent attack effects in both digital and physical attacks, exhibiting strong occlusion robustness. In addition of that, the adversarial background patches generated by this method achieve effective attacks in a smaller area and reduce the physical implementation cost.

In recent years, deep learning (DL) has achieved remarkable success in classifying hyperspectral images (HSIs), relying heavily on the quantity and quality of labeled samples. However, obtaining sufficient labels for HSIs poses a challenge. HSIs obtained by the same sensor often exhibit similar spectral information due to their shared physical, chemical properties, or reflective attributes. Joint analysis of several HSIs enables the integration of limited labeled samples and extraction of more robust and discriminative features from different HSIs. Therefore, a multitask collaborative network (MTCN) for the joint classification of multiple HSIs acquired by the same sensor in different areas is proposed. In the MTCN, each HSI has its own feature extraction channel, which facilitates the learning of image-specific representations. In addition, a feature sharing channel (FSC) is created to extract and transfer multihierarchical image-shared representations between multiple HSIs, thereby forming a common knowledge pool to facilitate feature sharing. Furthermore, a cross-channel mutual attention module (CMAM) is designed to collaboratively utilize features from image-specific and image-shared channels, enhancing the efficiency of information communication in HSIs. The experimental results on six HSIs demonstrate that the proposed MTCN can jointly classify multiple HSIs by the same sensor in different areas and achieve good classification performance.

Hyperspectral imaging has demonstrated substantial advantages in enhancing classification performance in remote sensing applications due to its abundant spectral information. To address the challenges of label noise and class imbalance in hyperspectral image (HSI) classification, we propose an end-to-end Feature-Guided Network (FGN) for HSI. Instead of merely combining spatial and channel attention, FGN leverages feature-level attention interactions to enhance contextual understanding, leading to better feature extraction, especially for underrepresented classes. Furthermore, a bidirectional loss for curriculum learning (CL) is proposed to rank the HSI training data in a descending or ascending order. The top and bottom loss regularizers are designed to make the proposed model suitable for noisy and imbalanced HSI data distributions. In the phase of selecting pace parameter, a stacking ensemble curriculum learning (SECL) model is established to avoid that the outliers and noisy HSI data are involved into the CL training process. A novel instruction matrix based on sample weights is designed for base classifiers. The outputs of the base models, combined with the expected labels, form the input-output pairs for training the second-level classifier. Experiments conducted on multiple hyperspectral imbalanced datasets with noisy labels demonstrate the superior performance of our method.

In the absence of sufficient labels, deep neural networks (DNNs) are prone to overfitting, resulting in poor performance and difficulty in training. Thus, many semisupervised methods aim to use unlabeled sample information to compensate for the lack of label quantity. However, as the available pseudolabels increase, the fixed structure of traditional models has difficulty in matching them, limiting their effectiveness. Therefore, a deep-growing neural network with manifold constraints (DGNN-MC) is proposed. It can deepen the corresponding network structure with the expansion of a high-quality pseudolabel pool and preserve the local structure between the original and high-dimensional data in semisupervised learning. First, the framework filters the output of the shallow network to obtain pseudolabeled samples with high confidence and adds them to the original training set to form a new pseudolabeled training set. Second, according to the size of the new training set, it increases the depth of the layers to obtain a deeper network and conducts the training. Finally, it obtains new pseudolabeled samples and deepens the layers again until the network growth is completed. The growing model proposed in this article can be applied to other multilayer networks, as their depth can be transformed. Taking HSI classification as an example, a natural semisupervised problem, the experimental results demonstrate the superiority and effectiveness of our method, which can mine more reliable information for better utilization and fully balance the growing amount of labeled data and network learning ability.

The precision of unsupervised point cloud registration methods is typically limited by the lack of reliable inlier estimation and self-supervised signal, especially in partially overlapping scenarios. In this article, we propose an effective inlier estimation method for unsupervised point cloud registration by capturing geometric structure consistency between the source point cloud and its corresponding reference point cloud copy. Specifically, to obtain a high-quality reference point cloud copy, a one-nearest neighborhood (1-NN) point cloud is generated by input point cloud, which facilitates matching map construction and allows for integrating dual neighborhood matching scores of 1-NN point cloud and input point cloud to improve matching confidence. Benefiting from the high-quality reference copy, we argue that the neighborhood graph formed by inlier and its neighborhood should have consistency between source point cloud and its corresponding reference copy. Based on this observation, we construct transformation-invariant geometric structure representations and capture geometric structure consistency to score the inlier confidence for estimated correspondences between source point cloud and its reference copy. This strategy can simultaneously provide the reliable self-supervised signals for model optimization. Finally, we further calculate transformation estimation by the weighted SVD algorithm with the estimated correspondences and the corresponding inlier confidence. We train the proposed model in an unsupervised manner, and extensive experiments on synthetic and real-world datasets illustrate the effectiveness of the proposed method.

The vulnerability to poor local optimum and the memorization of noise data limit the generalizability and reliability of massively parameterized convolutional neural networks (CNNs) on complex real-world data. Self-paced curriculum learning (SPCL), which models the easy-to-hard learning progression from human beings, is considered as a potential savior. In spite of the fact that numerous SPCL solutions have been explored, it still confronts two main challenges exactly in solving deep networks. By virtue of various designed regularizers, existing weighting schemes independent of the learning objective heavily rely on the prior knowledge. In addition, alternative optimization strategy (AOS) enables the tedious iterative training procedure, thus there is still not an efficient framework that integrates the SPCL paradigm well with networks. This article delivers a novel insight that attention mechanism allows for adaptive enhancement in the contribution of diverse instance information to the gradient propagation. Accordingly, we propose a general-purpose deep SPCL paradigm that incorporates the preferences of implicit regularizer for different samples into the network structure with inductive bias, which in turn is formalized as the self-paced curriculum network (SPCNet). Our proposal allows simultaneous online difficulty estimation, adaptive sample selection, and model updating in an end-to-end manner, which significantly facilitates the collaboration of SPCL to deep networks. Experiments on image classification and scene classification tasks demonstrate that our approach surpasses the state-of-the-art schemes and obtains superior performance.

DNN have demonstrated vulnerability to adversarial attacks in object detection tasks. While significant progress has been made in single-spectrum attacks, cross-spectral adversarial attacks remain challenging due to the complex tradeoffs between visible and infrared domains. To address this, an evolutionary multiobjective cross-spectral attack (MoXAttack) framework, for developing adversarial patches in closed-box cross-spectral scenarios is proposed. MoXAttack incorporates a multipopulation constraint-handling technique, which uses both penalty functions and feasibility rules to guide the search process. Spectrum-aware genetic operators are introduced to enhance solution diversity and feasibility. The framework automatically optimizes the smooth to cross-spectral shared patch shape using curvature energy. In addition, MoXAttack utilizes SVD for visible spectrum texture perturbations and adjustable thermal shielding material thickness for infrared spectrum control. Experiments on the LLVIP dataset demonstrate that MoXAttack achieves competitive performance across multiple object detection models. Ablation studies reveal the positive impact of improved components on attack effectiveness. The multipatch strategy improves attack success rates by at least 17%, while optimized patch shapes outperform conventional geometric shapes by at least 25% in terms of mAP drop. In the physical world test, the proposed method shows stability in different viewing angles.

This article proposes a novel evolutionary multitasking optimization (EMTO) paradigm called divide-and-conquer EMTO, which divides the original complex optimization problem into multiple simple optimization tasks and then these tasks are optimized by EMTO concurrently to formulate the resulting solution of the original problem. The main characteristics of divide-and-conquer EMTO are that the considered problem can be divided into multiple small-scale optimization tasks and the optimal solution is the combination of solutions of all tasks. In order to achieve the optimal combined fitness of all tasks, a relative improvement function and an adaptive exploration optimization strategy are designed for dynamic resource allocation across tasks. Finally, a case study on hyperspectral unmixing is investigated in the proposed divide-and-conquer EMTO framework by dividing the hyperspectral image into several homogeneous regions to formulate multiple sparse unmixing tasks. Experiments on benchmark and sparse unmixing problems demonstrate the superiority of divide-and-conquer EMTO.

This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton’s gradient update in large-scale stochastic optimization of machine learning models. It can be viewed as a variant of natural gradient descent, where the challenge of storing and calculating the full FIM is addressed through making use of the regularized FIM and directly finding the gradient update direction via Sherman-Morrison matrix inversion. Additionally, like the popular Adam method, SOFIM uses the first moment of the gradient to address the issue of non-stationary objectives across mini-batches due to heterogeneous data. The utilization of the regularized FIM and Sherman-Morrison matrix inversion leads to the improved convergence rate with the same space and time complexities as stochastic gradient descent (SGD) with momentum. The extensive experiments on training deep learning models using several benchmark image classification datasets demonstrate that the proposed SOFIM outperforms SGD with momentum and several state-of-the-art Newton optimization methods in term of the convergence speed for achieving the pre-specified objectives of training and test losses as well as test accuracy.

State-of-the-art (SOTA) adversarial attacks expose vulnerabilities in object detectors, often resulting in erroneous predictions. However, existing adversarial attacks neglect the stealth and flexibility of adversarial examples, which are crucial for conducting contextually consistent and inconspicuous attacks. To address these issues, leveraging the observed phenomenon of predicted box offsets in real-world object detection scenarios, this paper presents a novel adversarial attack framework called ShiftAttack. It leverages the concept of dense detection in prevalent object detectors, by boosting the confidence of low Intersection over Union (IoU) predictions within the positive samples (the set of predicted boxes responsible for localizing the same target), which leads to the erroneous exclusion of true positive predictions during the post-processing stage. Such a paradigm is highly stealthy as the shifted predictions seem like natural detector mistakes rather than obvious manipulations. To enhance the flexibility of ShiftAttack this paper proposes a generative approach called ShiftAttack Generator (SAG), which can not only shift predicted boxes for any target in arbitrary directions and distances but also facilitate adaptive feature exchange between pre- and post-shift regions to optimize the attack. Additionally, the proposed SAG incorporates the Dynamic Hinge Loss (DHL) to ensure the imperceptibility of perturbations, effectively mitigating the Patch-Pattern associated with the use of$\mathcal {L}_{2}$norm. Extensive experiments confirm that SAG surpasses other SOTA adversarial attacks in effectiveness, speed and stealthiness.

Multiparty learning provides solutions for training joint models with decentralized data under legal and practical constraints. However, traditional multiparty learning approaches are confronted with obstacles, such as system heterogeneity, statistical heterogeneity, and incentive design. Determining how to deal with these challenges and further improve the efficiency and performance of multiparty learning has become an urgent problem to be solved. In this article, we propose a novel contrastive multiparty learning framework for knowledge refinement and sharing with an accountable incentive mechanism. Since the existing parameter averaging method is contradictory to the learning paradigm of neural networks, we simulate the process of human cognition and communication and analogize multiparty learning as a many-to-one knowledge-sharing problem. The approach is capable of integrating the acquired explicit knowledge of each client in a transparent manner without privacy disclosure, and it reduces the dependence on data distribution and communication environments. The proposed scheme achieves significant improvement in model performance in a variety of scenarios, as we demonstrated through experiments on several real-world datasets.

Evolutionary multitasking optimization (EMTO) is a new optimization paradigm proposed in the field of evolutionary computation in recent years. EMTO can solve several different optimization tasks simultaneously and facilitate superior convergence characteristics by transferring effective knowledge among the tasks. However, existing EMTO usually focuses only on facilitating convergence characteristics while neglecting the potential privacy leakage problem in the knowledge transfer between different tasks. The privacy leakage could result in considerable financial losses or severe reputation impairment, which may impede the development of EMTO in real-world applications. To solve the problem of privacy leakage in EMTO, this paper proposes a privacy-enhanced multitasking particle swarm optimization algorithm. A knowledge transfer strategy with privacy preservation is designed based on homomorphic encryption by combining multitasking particle swarm optimization. In addition, an inter-task knowledge transfer mechanism implemented in a low-dimensional subspace is introduced to reduce the extra computational burden caused by privacy preservation. Comprehensive experiments are conducted on synthetic and NAS problems to verify the effectiveness of the proposed method. According to the experimental results, the proposed method has remarkable advantages in privacy preservation compared to existing EMTO.

Block sparse signals have the characteristics of nonzero values concentrated as blocks. Therefore, block sparse recovery requires recovering the original signal from the observed signal and the measurement matrix, where the recovered error should be small and the signal satisfies the block sparsity. In this article, block sparse recovery is solved as a multiobjective problem (MOP) and the recovery error, sparsity, and the block number of the recovered signal are considered as the conflicting objectives. Furthermore, the dimensionality of real block sparse signals is often too large, which increases the difficulty of recovery. Cooperative coevolutionary (CC) can effectively alleviate the curse of dimensionality in block sparse recovery. In addition, block sparsity causes zero and nonzero variables in the signal to form natural clusters, which is similar to the decomposition and cooperation of subproblems in CC. Therefore, CC is used in the algorithm to solve the block sparse MOP and decomposes the original problem into different subproblems. Each subproblem has a defined nonzero degree that can be combined with the proposed adaptive operator to adaptively adjust the way of generating new solutions of subproblem, which encourages the complete solution to have block sparsity. For the increased number of function evaluations (FEs) caused by the decomposition, a new list-inquiry evaluation is designed to avoid repeated evaluation. Finally, the experimental results on simulated data and real data have demonstrated the effectiveness of the proposed algorithm.

Multiobjective neural architecture search (MONAS) methods based on evolutionary algorithms (EAs) are inefficient when the evaluation of each architecture incorporates parameter learning from scratch. A surrogate-assisted MONAS problem can be tough considering cold-start in surrogate construction, and the evaluation of predicted promising architectures could still be cumbersome. Previously solved MONAS problems are likely to convey useful knowledge that could assist solving the current MONAS problem. To take the benefit from knowledge of these previous practices, a framework tackling large-scale knowledge transfer is proposed. Through sparse-constraint transfer stacking, the surrogate for the current problem gets informative easily. With knee-region knowledge distillation from previously learned parameters of nondominated architectures, evaluation of current architectures could be efficient and credible. To avoid transferring knowledge from irrelevant problems, an iterative source selection algorithm is designed to avoid negative transfer. The proposed framework is analyzed under different source and target MONAS problem combinations. Results show that with the help of this framework, architectures with competitive performance could be found under limited evaluation budget.

Point cloud registration is an important task in computer vision, where the goal is to estimate a transformation to align a pair of point clouds. Most of the existing registration methods face the problems of poor robustness and getting stuck in local optima. Evolutionary multitasking is an effective paradigm to enhance global search capability and improve convergence characteristics through knowledge transfer across multiple related tasks. Inspired by evolutionary multitasking, this article proposes a multiform optimization approach through evolutionary multitasking for solving the point cloud registration problems. We first construct two related registration tasks with different functional landscapes to form a multiform optimization problem. Compared with methods that only focus on a single registration attribute, the two proposed tasks focus on robustness and precision of registration, respectively. Then, a new two-stage bidirectional knowledge transfer strategy is presented, which can implement efficient knowledge transfer among two related tasks. Finally, both simulations and real experiments show the power of our method. The proposed method is robust to noise, outliers, and partial overlaps and is effective in multiple real registration scenarios, such as object registration, scene reconstruction, and simultaneous localization and mapping.

Classification of multitemporal remote sensing (MTRS) images with only a few labels of one of these images has attracted widespread interest in recent years. It usually confronts three problems: domain increment, class increment, and class disappearance. In this article, a robust self-paced incremental learning (RSPIL) is proposed to alleviate the above problems. First, change detection is used to transfer labels from the source image to the target one for formulating a combined training set. Then, the weighted classification loss and the distillation loss are considered to ensure classification performance and minimal forgetting. In particular, a novel entropy-inhibit loss is proposed to suppress the classification capability for the disappearing classes. These losses are combined with self-paced learning (SPL) by introducing a weight variable to measure the “easiness” of the training samples, which is able to automatically acquire accurate decision boundaries from easy to hard since the combined training set generated by change detection contains noisy samples and outliers. Finally, a nearest-average-eigenvector classifier and an exemplar set management strategy based on the sample weights are designed to alleviate catastrophic forgetting (CF). Twenty-four MTRS image datasets from four areas are considered in the experiments. The classification results demonstrate that the proposed method is able to alleviate CF and achieves significant improvements on multitemporal image datasets.

Recently, the field of hyperspectral image (HSI) classification has witnessed advancements with the emergence of deep learning models. Promising approaches, such as self-supervised strategies and domain adaptation, have effectively tackled the overfitting challenges posed by limited labeled samples in HSI classification. To extract comprehensive semantic information from different types of auxiliary tasks, which view the problem from multiple perspectives, and efficiently integrate multiple tasks into a single network, this paper proposes a hybrid multi-task learning framework (HyMuT) by sharing representations across multiple tasks. Based on the similarity between the data and target classification task, we construct three auxiliary tasks that are similar, related and weakly correlated to the target task, while three corresponding multi-task learning methods are integrated. The framework establishes a backbone network with a hard parameter sharing mechanism, which handles the main task and a similar spatial mask classification task. Subsequently, a hierarchical transfer multi-task learning approach is introduced to transfer the knowledge of a spatial-spectral joint mask reconstruction task from the autoencoder to the backbone network. Furthermore, a new source domain HSI dataset is introduced as an auxiliary task weakly correlated. To solve the source domain classification task and assist the hard parameter sharing mechanism, a dual adversarial classifier based on adversarial learning is employed. This classifier effectively extracts domain and task invariance. Extensive experiments are conducted on four benchmark HSI datasets to evaluate the performance. The results demonstrate that HyMuT outperforms state-of-the-art methods. This code will be available from the website: https://github.com/HaoLiu-XDU/HyMuT.

Remote sensing scene classification (RSSC) aims to understand and analyze the semantic information at the scene level with complex geographical properties. Despite the profound success of advanced deep models in automatically capturing hierarchical embedding representations and the gradual dominant trend in RSSC, it still remains a great challenge to precisely focus on targets at variable scales that are considered highly relevant to the corresponding scene and separated from the background. Motivated by this recognition, in this article, we present the gradient-guided multiscale focal attention network (GMFANet) for RSSC to adaptively localize the representative multiscale semantic representation for complex scenes. In particular, a lightweight parameterized hierarchical multiscale attention (HMA) mechanism is proposed, which constitutes the main aim of adaptively enhancing physical detail and high-level semantic information at different layers, rather than regarding each scale set with equivalent insight, while eliminating redundant information inherent in conventional attention mechanisms. Subsequently, a gradient-guided spatial focused attention (GSFA) module is specifically designed to accurately localize critical regions at multiple scales, with the dynamic combination of gradient-activated reference attention map and prediction attention map from supervised information-based learning. In addition, a curriculum-driven dynamic attention fusion (CDAF) strategy is tailored to fuse the spatial attention above from easy to hard for avoiding from poor local optimum and decreasing the early learning ambiguity. Our extensive comparative experiments and ablation analyses implemented on real-world public RSSC datasets indicate that our approach achieves the state-of-the-art performance exactly. The code is available athttps://github.com/bling2beyond/GMFANet.

Change detection (CD) in multispectral remote sensing (RS) imagery suffers from low spectral resolution which can lead to degraded recognition of change information from land cover objects. Considering that natural hyperspectral imagery (HSI) is much higher in spectral resolution and more accessible, using it to enhance the spectral information of RS multispectral imagery for CD can improve performance. To achieve this, we propose a spectral knowledge transfer (SKT) framework to allow the creation of pseudo-hyperspectral RS images from the available RS multispectral ones without the need for the real pairs of RS multispectral and hyperspectral images, typically required by existing RS spectral enhancement methods. Specifically, an autoencoder is first trained based on the available pairs of natural HSI and its multispectral counterparts and then calibrated via the available RS multispectral images. The finally obtained decoder module is used to generate the pseudo-hyperspectral image from an input RS multispectral image. We further propose a multispectrum collaborative CD (MCCD) framework that leverages both the real multispectral images and the pseudo hyperspectral images generated from them in a collaborative way to achieve performance improvement. Extensive experiments on two large-scale RS CD datasets and eight existing deep learning-based CD methods demonstrate the stronger efficacy of the proposed method.

Federated learning has gained prominence as an effective solution for addressing data silos, enabling collaboration among multiple parties without sharing their data. However, existing federated learning algorithms often neglect the challenge posed by multi-modal data distribution. Moreover, previous pioneering work face limitations in encrypting the exponential and logarithmic operations of the objective function with multiple independent variables, and they rely on a third-party cooperator for encryption. To address these limitations, this paper introduces a universal multi-modal vertical federated learning framework. To tackle the data distribution challenge, we propose a two-step multi-modal transformer model that captures cross-domain semantic features effectively. For encryption, where traditional additively homomorphic encryption algorithms fall short by supporting only addition and multiplication, we employ bivariate Taylor series expansion to transform the objective function. Integrating these components, we present a comprehensive training and transmission protocol that eliminates the need for a third-party cooperator during the encryption process. Extensive experiments conducted on diverse video-text and image-text datasets validate the superior performance of our framework compared to state-of-the-art approaches, affirming its effectiveness in multi-modal vertical federated learning settings.

The way in which autonomous transport will be adopted is likely to determine the net social benefits delivered by the technology and the sustainability of the transport system. Autonomous vehicles (AVs) will change travel behavior due to reduction in the effort needed for humans to drive a vehicle, the need for them to find a parking space, and the costs related to vehicle operation. The AVs’ benefits are likely to increase their adoption compared to conventional human-driven vehicles, possibly leading to more vehicle kilometers travelled (VKT) and consequently weakening their benefits in large cities particularly if they are used in competition with public transport (PT). This paper evaluates the interplay between AVs and PT, and how road network pricing can be used to influence behavioral changes when personal autonomous vehicles (PAVs) are highly available. An agent-based demand model framework is proposed to estimate the mode share of PAVs and PT based on their perceived travelling costs on a real transport network in Melbourne, Australia. The modelling results suggest that convenient and affordable PAVs could compete with traditional PT and reduce overall PT patronage by up to 10%. However, through considering road network pricing schemes, the role of PAVs could be shifted from competing with PT to a complementary first-and-last-mile service that increases PT share by almost 17%. The results also show that road pricing policies can be used as effective interventions to manage PAV operations by reducing empty vehicle trips by 20%.

Federated Learning (FL) achieves great popularity in the Internet of Things (IoT) as a powerful interface to offer intelligent services to customers while maintaining data privacy. Under the orchestration of a server, edge devices (also called clients in FL) collaboratively train a global deep-learning model without sharing any local data. Nevertheless, the unequal training contributions among clients have made FL vulnerable, as clients with heavily biased datasets can easily compromise FL by sending malicious or heavily biased parameter updates. Furthermore, the resource shortage issue of the network also becomes a bottleneck. Due to overwhelming computation overheads generated by training deep-learning models on edge devices, and significant communication overheads for transmitting deep-learning models across the network, enormous amounts of resources are consumed in the FL process. This encompasses computation resources like energy and communication resources like bandwidth. To comprehensively address these challenges, in this paper, we present FLrce, an efficient FL framework with arelationship-basedclient selection andearly-stopping strategy. FLrce accelerates the FL process by selecting clients with more significant effects, enabling the global model to converge to a high accuracy in fewer rounds. FLrce also leverages an early stopping mechanism that terminates FL in advance to save communication and computation resources. Experiment results show that, compared with existing efficient FL frameworks, FLrce improves the computation and communication efficiency by at least 30% and 43% respectively.

Learning effective representations from unlabeled data is a challenging task for point cloud understanding. As the human visual system can map concepts learned from 2D images to the 3D world, and inspired by recent multimodal research, we introduce data from point cloud modality and image modality for joint learning. Based on the properties of point clouds and images, we propose CrossNet, a comprehensive intra- and cross-modal contrastive learning method that learns 3D point cloud representations. The proposed method achieves 3D-3D and 3D-2D correspondences of objectives by maximizing the consistency of point clouds and their augmented versions, and with the corresponding rendered images in invariant space. We further distinguish the rendered images into RGB and grayscale images to extract color and geometric features, respectively. These training objectives combine feature correspondences between modalities to combine rich learning signals from point clouds and images. Our CrossNet is simple: we add a feature extraction module and a projection head module to the point cloud and image branches, respectively, to train the backbone network in a self-supervised manner. After the network is pretrained, only the point cloud feature extraction module is required for fine-tuning and directly predicting results for downstream tasks. Our experiments on multiple benchmarks demonstrate improved point cloud classification and segmentation results, and the learned representations can be generalized across domains.

The pandemic of coronavirus disease 2019 (COVID-19) has led to a global public health crisis, which caused millions of deaths and billions of infections, greatly increasing the pressure on medical resources. With the continuous emergence of viral mutations, developing automated tools for COVID-19 diagnosis is highly desired to assist the clinical diagnosis and reduce the tedious workload of image interpretation. However, medical images in a single site are usually of a limited amount or weakly labeled, while integrating data scattered around different institutions to build effective models is not allowed due to data policy restrictions. In this article, we propose a novel privacy-preserving cross-site framework for COVID-19 diagnosis with multimodal data, seeking to effectively leverage heterogeneous data from multiple parties while preserving patients' privacy. Specifically, a Siamese branched network is introduced as the backbone to capture inherent relationships across heterogeneous samples. The redesigned network is capable of handling semisupervised inputs in multimodalities and conducting task-specific training, in order to improve the model performance of various scenarios. The framework achieves significant improvement compared with state-of-the-art methods, as we demonstrate through extensive simulations on real-world datasets.

Three-dimensional point cloud registration is an important field in computer vision. Recently, due to the increasingly complex scenes and incomplete observations, many partial-overlap registration methods based on overlap estimation have been proposed. These methods heavily rely on the extracted overlapping regions with their performances greatly degraded when the overlapping region extraction underperforms. To solve this problem, we propose a partial-to-partial registration network (RORNet) to find reliable overlapping representations from the partially overlapping point clouds and use these representations for registration. The idea is to select a small number of key points called reliable overlapping representations from the estimated overlapping points, reducing the side effect of overlap estimation errors on registration. Although it may filter out some inliers, the inclusion of outliers has a much bigger influence than the omission of inliers on the registration task. The RORNet is composed of overlapping points' estimation module and representations' generation module. Different from the previous methods of direct registration after extraction of overlapping areas, RORNet adds the step of extracting reliable representations before registration, where the proposed similarity matrix downsampling method is used to filter out the points with low similarity and retain reliable representations, and thus reduce the side effects of overlap estimation errors on the registration. Besides, compared with previous similarity-based and score-based overlap estimation methods, we use the dual-branch structure to combine the benefits of both, which is less sensitive to noise. We perform overlap estimation experiments and registration experiments on the ModelNet40 dataset, outdoor large scene dataset KITTI, and natural data Stanford Bunny dataset. The experimental results demonstrate that our method is superior to other partial registration methods. Our code is available at https://github.com/superYuezhang/RORNet.

Multiparty learning (MPL) is an emerging framework for privacy-preserving collaborative learning. It enables individual devices to build a knowledge-shared model and remaining sensitive data locally. However, with the continuous increase of users, the heterogeneity gap between data and equipment becomes wider, which leads to the problem of model heterogeneous. In this article, we concentrate on two practical issues: data heterogeneous problem and model heterogeneous problem, and propose a novel personal MPL method named device-performance-driven heterogeneous MPL (HMPL). First, facing the data heterogeneous problem, we focus on the problem of various devices holding arbitrary data sizes. We introduce a heterogeneous feature-map integration method to adaptively unify the various feature maps. Meanwhile, to handle the model heterogeneous problem, as it is essential to customize models for adapting to the various computing performances, we propose a layer-wise model generation and aggregation strategy. The method can generate customized models based on the device's performance. In the aggregation process, the shared model parameters are updated through the rules that the network layers with the same semantics are aggregated with each other. Extensive experiments are conducted on four popular datasets, and the result demonstrates that our proposed framework outperforms the state of the art (SOTA).

In this modern era, traffic congestion has become a major source of negative economic and environmental impact for urban areas worldwide. One of the most efficient ways to mitigate this issue is through traffic prediction. This research field has evolved greatly ever since its inception in the late 70s. Recently, deep neural network models have gained popularity thanks to its predictive power, but despite this, literature surveys of such methods are rare; making it difficult to ascertain the progress of this research field. In this work, we address this issue by presenting an up-to-date survey of deep neural network for traffic prediction. We provide detailed explanations of popular deep neural network architectures used in the traffic flow prediction literatures, categorize and describe the literatures themselves, present an overview of the commonalities and differences among different works, and finally provide a discussion regarding the challenges and future directions for this field.

Physics-informed neural networks (PINN s) are a newly emerging research frontier in machine learning, which incorporate certain physical laws that govern a given data set, e.g., those described by partial differential equations (PDEs), into the training of the neural network (NN) based on such a data set. In PINN s, the NN acts as the solution approximator for the PDE while the PDE acts as the prior knowledge to guide the NN training, leading to the desired generalization performance of the NN when facing the limited availability of training data. However, training PINNs is a non-trivial task largely due to the complexity of the loss composed of both NN and physical law parts. In this work, we propose a new PINN training framework based on the multi-task optimization (MTO) paradigm. Under this framework, multiple auxiliary tasks are created and solved together with the given (main) task, where the useful knowledge from solving one task is transferred in an adaptive mode to assist in solving some other tasks, aiming to uplift the performance of solving the main task. We implement the proposed framework and apply it to train the PINN for addressing the traffic density prediction problem. Experimental results demonstrate that our proposed training framework leads to significant performance improvement in comparison to the traditional way of training the PINN.

Multi-party learning provides an effective approach for training a machine learning model, e.g., deep neural networks (DNNs), over decentralized data by leveraging multiple decentralized computing devices, subjected to legal and practical constraints. Different parties, so-called local participants, usually provide heterogenous data in a decentralized mode, leading to non-IID data distributions across different local participants which pose a notorious challenge for multi-party learning. To address this challenge, we propose a novel heterogeneous differentiable sampling (HDS) framework. Inspired by the dropout strategy in DNNs, a data-driven network sampling strategy is devised in the HDS framework, with differentiable sampling rates which allow each local participant to extract from a common global model the optimal local model that best adapts to its own data properties so that the size of the local model can be significantly reduced to enable more efficient inference. Meanwhile, co-adaptation of the global model via learning such local models allows for achieving better learning performance under non-IID data distributions and speeds up the convergence of the global model. Experiments have demonstrated the superiority of the proposed method over several popular multi-party learning techniques in the multi-party settings with non-IID data distributions.

Cloud platforms scheduling resources based on the demand of the tasks submitted by the users, is critical to the cloud provider's interest and customer satisfaction. In this paper, we propose a multi-objective cloud task scheduling algorithm based on an evolutionary multi-factorial optimization algorithm. First, we choose execution time, execution cost, and virtual machines load balancing as the objective functions to construct a multi-objective cloud task scheduling model. Second, the multi-factor optimization (MFO) technique is applied to the task scheduling problem, and the task scheduling characteristics are combined with the multi-objective multi-factor optimization (MO-MFO) algorithm to construct an assisted optimization task. Finally, a dynamic adaptive transfer strategy is designed to determine the similarity between tasks according to the degree of overlap of the MFO problem and to control the intensity of knowledge transfer. The results of simulation experiments on the cloud task test dataset show that our method significantly improves scheduling efficiency, compared with other evolutionary algorithms (EAs), the scheduling method simplifies the decomposition of complex problems by a multi-factor approach, while using knowledge transfer to share the convergence direction among sub-populations, which can find the optimal solution interval more quickly and achieve the best results among all objective functions.

The performance of machine learning algorithms heavily relies on the availability of a large amount of training data. However, in reality, data usually reside in distributed parties such as different institutions and may not be directly gathered and integrated due to various data policy constraints. As a result, some parties may suffer from insufficient data available for training machine learning models. In this article, we propose a multiparty dual learning (MPDL) framework to alleviate the problem of limited data with poor quality in an isolated party. Since the knowledge-sharing processes for multiple parties always emerge in dual forms, we show that dual learning is naturally suitable to handle the challenge of missing data, and explicitly exploits the probabilistic correlation and structural relationship between dual tasks to regularize the training process. We introduce a feature-oriented differential privacy with mathematical proof, in order to avoid possible privacy leakage of raw features in the dual inference process. The approach requires minimal modifications to the existing multiparty learning structure, and each party can build flexible and powerful models separately, whose accuracy is no less than nondistributed self-learning approaches. The MPDL framework achieves significant improvement compared with state-of-the-art multiparty learning methods, as we demonstrated through simulations on real-world datasets.

Graph classification aims to predict the label associated with a graph and is an important graph analytic task with widespread applications. Recently, graph neural networks (GNNs) have achieved state-of-the-art results on purely supervised graph classification by virtue of the powerful representation ability of neural networks. However, almost all of them ignore the fact that graph classification usually lacks reasonably sufficient labeled data in practical scenarios due to the inherent labeling difficulty caused by the high complexity of graph data. The existing semisupervised GNNs typically focus on the task of node classification and are incapable to deal with graph classification. To tackle the challenging but practically useful scenario, we propose a novel and general semisupervised GNN framework for graph classification, which takes full advantage of a slight amount of labeled graphs and abundant unlabeled graph data. In our framework, we train two GNNs as complementary views for collaboratively learning high-quality classifiers using both labeled and unlabeled graphs. To further exploit the view itself, we constantly select pseudo-labeled graph examples with high confidence from its own view for enlarging the labeled graph dataset and enhancing predictions on graphs. Furthermore, the proposed framework is investigated on two specific implementation regimes with a few labeled graphs and the extremely few labeled graphs, respectively. Extensive experimental results demonstrate the effectiveness of our proposed semisupervised GNN framework for graph classification on several benchmark datasets.

End-to-end deep clustering method utilizes deep neural networks to jointly learn representation features and clustering assignments. Although many k-means-friendly deep clustering models have been explored, the existing division-based methods tend to directly implement clustering with a specific number of clusters, which suffers from poor performance resulted from indistinguishable clusters, and contributes to bad local optimum. At the same time, the representation learning of fuzzy$c$-means clustering in the feature space still needs more research. In this article, a deep fuzzy curriculum clustering method with the learning strategy of clustering from easy to complex automatically is proposed to tackle the above issues. First, considering the soft flexible allocation of fuzzy$c$-means and the preservation of local structure of original data, the fuzzy clustering loss and the autoencoder's reconstruction loss are constructed to learn the embedded features and clustering centers simultaneously. Second, curriculum loss is introduced into the constraint to make clusters successively merge in line with implementing clustering from easy to complex, and realize the bottom-up deep aggregative clustering automatically. In addition, novel curriculum information is proposed as constraint to guide the merging of clusters belonging to the same class. Experimental results on four real-world datasets show the superiority of the proposal.

Classification of remote sensing image series which differ in quality and details, has impportant implications for the analysis of land cover, whereas it is expensive and time-consuming as a result of manual annotations. Fortunately, domain adaptation (DA) provides an outstanding solution to the problem. However, information loss while aligning two distributions often exists in traditional DA methods, which impacts the effect of classification with DA. To alleviate this issue, an inter-domain representation alignment and fine-tuning based network (RAFNet) is proposed for image series classification. Inter-domain representation alignment, which is fulfilled by a variational auto-encoder (VAE) trained by both source and target data, encourages reducing the discrepancy between the two marginal distributions of different domains and simultaneously preserving more data properties. As a result, RAFNet, which fuses the multi-scale aligned representations, performs classification task in the target domain after well trained with supervised learning in the source domain. Specifically, the multi-scale aligned representations of RAFNet is acquired by duplicating the frozen encoder of VAE. Then, an information based loss function is designed to fine-tune RAFNet, in which both the unchanged and changed information implied in change maps is completely used to learn the discriminative features better and make the model more generalized for the target domain. Finally, experiment studies on three datasets validate the effectiveness of RAFNet with considerable segmentation accuracy even the target data has no access to any annotated information.

Graph Neural Network (GNN) is capable of applying deep neural networks to graph domains. Recently, Message Passing Neural Networks (MPNNs) have been proposed to generalize several existing graph neural networks into a unified framework. For graph representation learning, MPNNs first generate discriminative node representations using the message passing function and then read from the node representation space to generate a graph representation using the readout function. In this paper, we analyze the representation capacity of the MPNNs for aggregating graph information and observe that the existing approaches ignore the self-loop for graph representation learning, leading to limited representation capacity. To alleviate this issue, we introduce a simple yet effective propagation enhanced extension, Self-Connected Neural Message Passing (SC-NMP), which aggregates the node representations of the current step and the graph representation of the previous step. To further improve the information flow, we also propose a Densely Self-Connected Neural Message Passing (DSC-NMP) that connects each layer to every other layer in a feed-forward fashion. Both proposed architectures are applied at each layer and the graph representation can then be used as input into all subsequent layers. Remarkably, combining these two architectures with existing GNN variants can improve these models’ performance for graph representation learning. Extensive experiments on various benchmark datasets strongly demonstrate the effectiveness, leading to superior performance for graph classification and regression tasks.

Graph neural networks (GNNs) have demonstrated great success in many graph data-based applications. The impressive behavior of GNNs typically relies on the availability of a sufficient amount of labeled data for model training. However, in practice, obtaining a large number of annotations is prohibitively labor-intensive and even impossible. Co-training is a popular semi-supervised learning (SSL) paradigm, which trains multiple models based on a common training set while augmenting the limited amount of labeled data used for training each model via the pseudolabeled data generated from the prediction results of other models. Most of the existing co-training works do not control the quality of pseudolabeled data when using them. Therefore, the inaccurate pseudolabels generated by immature models in the early stage of the training process are likely to cause noticeable errors when they are used for augmenting the training data for other models. To address this issue, we propose a self-paced co-training for the GNN (SPC-GNN) framework for semi-supervised node classification. This framework trains multiple GNNs with the same or different structures on different representations of the same training data. Each GNN carries out SSL by using both the originally available labeled data and the augmented pseudolabeled data generated from other GNNs. To control the quality of pseudolabels, a self-paced label augmentation strategy is designed to make the pseudolabels generated at a higher confidence level to be utilized earlier during training such that the negative impact of inaccurate pseudolabels on training data augmentation, and accordingly, the subsequent training process can be mitigated. Finally, each of the trained GNN is evaluated on a validation set, and the best-performing one is chosen as the output. To improve the training effectiveness of the framework, we devise a pretraining followed by a two-step optimization scheme to train GNNs. Experimental results on the node classification task demonstrate that the proposed framework achieves significant improvement over the state-of-the-art SSL methods.

We propose a data-driven distributed machine learning approach to scalably predict the trustworthiness of homogeneous IoT services in heterogeneous Mobile Edge Computing (MEC)-based IoT systems. The proposed approach formulates training distributed trust prediction models within an MEC-based IoT system as a Network Lasso problem. We then introduce a variant of Stochastic Alternating Method of Multipliers framework (S-ADMM) enriched with the ability for feature selection at each MEC layer. To verify the effectiveness of the proposed approach, we carried out a comprehensive evaluation on three real-world datasets adjusted to exhibit the context-dependent trust information accumulated in MEC environments within a given MEC topology. The experimental results affirmed the effectiveness of our approach and its suitability to predict trustworthiness of IoT services in MEC-based IoT systems.

Mobile Edge Computing (MEC)-based Internet of Things (IoT) systems generate trust information in a real-time and distributed manner. Predicting trustworthiness of IoT services in such an MEC environment requires new prediction strategies that cater for the aforementioned characteristics of trust information. More importantly, it is imperative to investigate how the real-time trust information could be effectively integrated into trust prediction strategies in order to capture the ever-evolving nature of trustworthiness of IoT services. In turn, such a strategy allows IoT service consumers to derive more relevant and accurate trust-based decisions. To that end, our work models trust prediction in MEC-based IoT systems as an online regularized finite-sum problem in a distributed MEC environment with a given MEC topology. We then adopt the Online Alternating Direction Method (OADM) to effectively train trust prediction models in parallel over the distributed MEC environment. OADM allows splitting the aforementioned finite-sum problem into multiple sub-problems that correspond to different local MEC environments. These sub-problems can then be solved iteratively within each local MEC environment by using the local trust data therein. This can avoid the movement of data across the core networks of mobile network providers. Experiments on real-world and synthetic datasets demonstrate the effectiveness and scalability of the proposed method.

Deep neural networks (DNNs) often rely on massive labelled data for training, which is inaccessible in many applications. Data augmentation (DA) tackles data scarcity by creating new labelled data from available ones. Different DA methods have different mechanisms and therefore using their generated labelled data for DNN training may help improving DNN's generalisation to different degrees. Combining multiple DA methods, namely multi-DA, for DNN training, provides a way to further boost generalisation. Among existing multi-DA based DNN training methods, those relying on knowledge distillation (KD) have received great attention. They leverage knowledge transfer to utilise the labelled data sets created by multiple DA methods instead of directly combining them for training DNNs. However, existing KD-based methods can only utilise certain types of DA methods, incapable of making full use of the advantages of arbitrary DA methods. In this work, we propose a general multi-DA based DNN training framework capable to use arbitrary DA methods. To train a DNN, our framework replicates a certain portion in the latter part of the DNN into multiple copies, leading to multiple DNNs with shared blocks in their former parts and independent blocks in their latter parts. Each of these DNNs is associated with a unique DA and a newly devised loss that allows comprehensively learning from the data generated by all DA methods and the outputs from all DNNs in an online and adaptive way. The overall loss, i.e., the sum of each DNN's loss, is used for training the DNN. Eventually, one of the DNNs with the best validation performance is chosen for inference. We implement the proposed framework by using three distinct DA methods and apply it for training representative DNNs. Experimental results on the popular benchmarks of image classification demonstrate the superiority of our method to several existing single-DA and multi-DA based training methods.

Real-world electricity consumption prediction may involve different tasks, e.g., prediction for different time steps ahead or different geo-locations. These tasks are often solved independently without utilizing some common problem-solving knowledge that could be extracted and shared among these tasks to augment the performance of solving each task. In this work, we propose a multi-task optimization (MTO) based co-training (MTO-CT) framework, where the models for solving different tasks are co-trained via an MTO paradigm in which solving each task may benefit from the knowledge gained from when solving some other tasks to help its solving process. MTO-CT leverages long short-term memory (LSTM) based model as the predictor where the knowledge is represented via connection weights and biases. In MTO-CT, an inter-task knowledge transfer module is designed to transfer knowledge between different tasks, where the most helpful source tasks are selected by using the probability matching and stochastic universal selection, and evolutionary operations like mutation and crossover are performed for reusing the knowledge from selected source tasks in a target task. We use electricity consumption data from five states in Australia to design two sets of tasks at different scales: a) one-step ahead prediction for each state (five tasks) and b) 6-step, 12-step, 18-step, and 24-step ahead prediction for each state (20 tasks). The performance of MTO-CT is evaluated on solving each of these two sets of tasks in comparison to solving each task in the set independently without knowledge sharing under the same settings, which demonstrates the superiority of MTO-CT in terms of prediction accuracy.

Deep convolutional neural networks (CNNs) have achieved tremendous successes but tend to suffer from high computation costs mainly due to heavy over-parameterization, resulting in the difficulty of directly applying them to the ever-growing application demands based on low-end edge devices with strong power restriction and real-time inference requirement. Recently, there has much research attention devoted to compressing the network via pruning to address this issue. Most of the existing methods rely on some hand-designed pruning rules, which suffer from several limitations. Firstly, manually designed rules are only applicable to limited application scenarios, which can hardly generalize well in a broader scope. And these rules are typically designed based on human experience and via trial and error, and thus highly subjective. Then, channels of different layers in a network may have diverse distributions, which means the same pruning rule is not appropriate for each layer. To address these limitations, we propose a novel channel pruning scheme, in which the task-irrelevant channels are removed in a task-driven manner. Specifically, an adaptively differentiable search module is proposed to find the best pruning rule automatically for different layers in CNNs under sparsity constraints. Besides, we employed knowledge distillation to alleviate the excessive performance loss. Once the training process is finished, a compact network will be obtained by removing channels based on layer-wise pruning rules. We have evaluated the proposed method on some well-known benchmark datasets including CIFAR, MNIST, and ImageNet in comparison to several state-of-the-art pruning methods. Experimental results demonstrate the superiority of our method over the compared ones in terms of both parameters and FLOPs reduction.

Deep learning-based image inpainting has been widely studied, leading to great success. However, many methods adopt convolution and normalization operations, which will bring up some issues to affect the performance. The vanilla normalization cannot distinguish the pixels in corrupted regions from the other valid pixels, resulting in the mean and variance shifts. In addition, the limited receptive field of convolution makes it unable to capture long-range valid information directly. In order to tackle these challenges, we propose a novel deep generative model for image inpainting with two key modules, namely, the channel and spatially adaptive batch normalization (CSA-BN) module, and the selective latent-space-mapping-based contextual attention (SLSM-CA) layer. We replace the vanilla normalization with the CSA-BN module. By channel and spatially adaptive denormalization, the CSA-BN module can mitigate the spatial mean and variance shifts in each channel in a targeted way. In addition, we also integrate the SLSM-CA layer into our model to capture the long-range correlations explicitly. By introducing dual-branch attention and a feature selection module, the SLSM-CA layer can selectively utilize the multi-scale background information to improve prediction quality. What’s more, it introduces the latent spaces to achieve the low-rank approximations of attention matrices and to reduce computational costs. Extensive quantitative and qualitative evaluations demonstrate the superiority of the proposed method compared with state-of-the-art methods.

Detecting anomalies in videos is a fundamental issue in public security. The majority of existing deep learning methods often perform anomaly detection based on the behavior or the trajectory of a single target. However, due to the overlaps of the crowd and the low-resolution of monitoring images, the segmentation of population is hard to implement and the features cannot be learned thoroughly, which make the methods be easily disturbed by visual elements and thus may lead to false detection sometimes. To tackle these problems, we propose the influence-aware attention to learn the representative attributes of the whole crowd. Walking pedestrians can be divided into numbers of flows, and in this paper, we aim to measure the consistency of movement patterns in the same stream and the interactions between different streams. Meanwhile, great importance is given to the relation between pedestrians and the circumstance for certain anomalies occur as a result of environmental issues. Specifically, the influence-aware attention module is composed of the motion attention and the location attention, which is designed to quantify the relations in the scene from spatial and temporal aspects. For the lack of abnormal samples, we utilize a dual generator-based framework to learn interactions among normal scenes. Experimental results on six benchmarks verify the effectiveness and robustness of our proposed method.

The searching ability of the population-based search algorithms strongly relies on the coordinate system on which they are implemented. However, the widely used coordinate systems in the existing multifactorial optimization (MFO) algorithms are still fixed and might not be suitable for various function landscapes with differential modalities, rotations, and dimensions; thus, the intertask knowledge transfer might not be efficient. Therefore, this article proposes a novel intertask knowledge transfer strategy for MFOs implemented upon an active coordinate system that is established on a common subspace of two search spaces. The proper coordinate system might identify some common modality in a proper subspace to some extent. In this article, to seek the intermediate subspace, we innovatively introduce the geodesic flow that starts from a subspace, reaching another subspace in unit time. A low-dimension intermediate subspace is drawn from a uniform distribution defined on the geodesic flow, and the corresponding coordinate system is given. The intertask trial generation method is applied to the individuals by first projecting them on the low-dimension subspace, which reveals the important invariant features of the multiple function landscapes. Since intermediate subspace is generated from the major eigenvectors of tasks' spaces, this model turns out to be intrinsically regularized by neglecting the minor and small eigenvalues. Therefore, the transfer strategy can alleviate the influence of noise led by redundant dimensions. The proposed method exhibits promising performance in the experiments.

Evolutionary multitasking (EMT), which shares knowledge across multiple tasks while the optimization progresses online, has demonstrated superior performance in terms of both optimization quality and convergence speed over its single-task counterpart in solving complex optimization problems. However, most of the existing EMT algorithms only consider handling two tasks simultaneously. As the computational cost incurred in the evolutionary search and knowledge transfer increased rapidly with the number of optimization tasks, these EMT algorithms cannot meet today’s requirements of optimization service on the cloud for many real-world applications, where hundreds or thousands of optimization requests (labeled as large-scale EMT) are often received simultaneously and require to be optimized in a short time. Recently, graphics processing unit (GPU) computing has attracted extensive attention to accelerate the applications possessing large-scale data volume that are traditionally handled by the central processing unit (CPU). Taking this cue, toward large-scale EMT, in this article, we propose a new EMT paradigm based on the island model with the compute unified device architecture (CUDA), which is able to handle a large number of continuous optimization tasks efficiently and effectively. Moreover, under the proposed paradigm, we develop the GPU-basedimplicitandexplicitknowledge transfer mechanisms for EMT. To evaluate the performance of the proposed paradigm, comprehensive empirical studies have been conducted against its CPU-based counterpart in large-scale EMT.

Evolutionary multitask optimization (EMTO) studies how to simultaneously solve multiple optimization tasks via evolutionary algorithms (EAs) while making the useful knowledge acquired from solving one task to assist solving other tasks, aiming to improve the overall performance of solving each individual task. Recent years have seen a large body of EMTO works based on different kinds of EAs and studying one or more aspects in how to represent, extract, transfer, and reuse knowledge. A key challenge to EMTO is the occurrence of negative knowledge transfer between tasks, which becomes severer when the total number of tasks increases. To address this issue, we propose an adaptive EMTO (AEMTO) framework. This framework can adapt knowledge transfer frequency, knowledge source selection, and knowledge transfer intensity in a synergistic way to make the best use of knowledge transfer, especially when facing many tasks. We implement the proposed AEMTO framework and evaluate our implementation on three suites of MTO problems with 2, 10, and 50 tasks and one real-world MTO problem with 2000 tasks in comparison to several state-of-the-art EMTO methods with certain adaptation strategies regarding knowledge transfer and the single-task optimization counterpart of the proposed method. Experimental results have demonstrated the effectiveness of the adaptive knowledge transfer strategies used in AEMTO and the overall performance superiority of AEMTO.

Hyperspectral images (HSIs) contain rich spectral signatures that reveal more image details and, thus, enable the detection of less noticeable changes on the ground. However, HSI-based change detection (CD) is susceptible to a large amount of irrelevant or noisy spectral and spatial information due to massive spectral bands. To address these issues, we propose a novel spectral and spatial attention network (S2AN) for HSI-based CD, which is capable to suppress CD-irrelevant spectral and spatial information via adaptive spectral and spatial attention mechanisms. S2AN takes as input the image patch from the difference map between two HSIs and outputs the status of change for the patch. Specifically, S2AN is composed of several repeated attention blocks, each of which contains the spectral attention (SpeA) module for directly calculating the attention score for each input channel, the Gaussian spatial attention (GSpaA) module that first constructs an adaptive Gaussian distribution and then samples it to derive the attention scores for each spatial position, and the convolutional feature extraction (CFE) module for extracting features from the attention-weighted input. It is worth mentioning that, in addition to the advantage of the attention, GSpaA also reduces the sensitivity of patch size for patch-based methods. To effectively train S2AN when facing insufficient labeled data, a semisupervised strategy that combines supervised and unsupervised methods to augment labeled training data is proposed. Experiments on several HSI datasets in comparison to existing methods show the superiority of S2AN.

Nowadays, due to the difficult acquisition of true labels, a semisupervised neural network has shown great potential for change detection (CD) in remote sensing images. However, most of the traditional semisupervised neural network detection frameworks are complex to train and require additional structural analysis, along with a fixed structure, lacking universality. In this article, a semisupervised adaptive ladder network (SSALN) for remote sensing image CD is proposed, which enables dual-input label-incremental architecture searching with a concise and variable structure. First, SSALN is suitable for CD from two remote sensing images of any type with the characteristic of minimal label dependency and automatic network structure adjustment. The network can generate more reliable pseudolabels through continuous iterations to help limited real labels exploit implicit information, identify the most effective network, and form the ascending network structure optimization. Second, the acquisition of pseudolabels is the fusion of semisupervised and unsupervised CD approaches, which ensures the multiperspective information supplement. Multiple CD maps are fused to generate labels for the next iteration, making the predicting more reliable. Finally, both homogenous images and heterogenous images are tested with experiments. Even if the detection object is switched, it can be well adaptive and compatible without manual modification of the network. Experimental results demonstrate that the proposed method can promote the flow of label information through structure searching and self-circulation in the ascending network optimization; thus, it has outstanding performance on tasks of remote sensing image CD.

Accurate house prediction is of great significance to various real estate stakeholders such as house owners, buyers, and investors. We propose a location-centered prediction framework that differs from existing work in terms of data profiling and prediction model. Regarding data profiling, we make an important observation as follows – besides the in-house features such as floor area, the location plays a critical role in house price prediction. Unfortunately, existing work either overlooked it or had a coarse grained measurement of locations. Thereby, we define and capture a fine-grained location profile powered by a diverse range of location data sources, including transportation profile, education profile, suburb profile based on census data, and facility profile. Regarding the choice of prediction model, we observe that a variety of approaches either consider the entire data for modeling, or split the entire house data and model each partition independently. However, such modeling ignores the relatedness among partitions, and for all prediction scenarios, there may not be sufficient training samples per partition for the latter approach. We address this problem by conducting a careful study of exploiting the Multi-Task Learning (MTL) model. Specifically, we map the strategies for splitting the entire house data to the ways the tasks are defined in MTL, and select specific MTL-based methods with different regularization terms to capture and exploit the relatedness among tasks. Based on real-world house transaction data collected in Melbourne, Australia, we design extensive experimental evaluations, and the results indicate a significant superiority of MTL-based methods over state-of-the-art approaches. Meanwhile, we conduct an in-depth analysis on the impact of task definitions and method selections in MTL on the prediction performance, and demonstrate that the impact of task definitions on prediction performance far exceeds that of method selections.

Representing nodes in a network as low-dimensional dense vectors can facilitate the analysis of complex networks, which is a challenging task and has attracted increasing attention. However, in the real world, networks are changing over time, such as cooperation in citation networks and communication in email networks. Most of the recent embedding methods only focus on static networks. Thus they ignore the critical temporal information, which serves as a supplement to structure information and has been proved to improve the quality of node embedding. In this work, we propose an unsupervised deep learning model called DTINE, which explores temporal information for further enhancing the robustness of node representations in dynamic networks. To preserve network topology, we pertinently design a temporal weight and sampling strategy to extract features from the neighborhoods. An attention mechanism will be applied on the recurrent neural network to measure the contributions of historical information and capture the evolution of the networks. Experimental results on four real-world networks demonstrate that the proposed method achieves better performance than state-of-the-art methods.

In this modern era, traffic congestion has become a major source of severe negative economic and environmental impact for urban areas worldwide. One of the most efficient ways to mitigate traffic congestion is through future traffic prediction. The research field of traffic prediction has evolved greatly ever since its inception in the late 70s. Earlier studies mainly use classical statistical models such as ARIMA and its variants. Recently, researchers have started to focus on machine learning models because of their power and flexibility. As theoretical and technological advances emerge, we enter the era of deep neural network, which gained popularity due to its sheer prediction power which can be attributed to the complex and deep structure. Despite the popularity of deep neural network models in the field of traffic prediction, literature surveys of such methods are rare. In this work, we present an up-to-date survey of deep neural network for traffic prediction. We will provide a detailed explanation of popular deep neural network architectures commonly used in the traffic flow prediction literatures, categorize and describe the literatures themselves, present an overview of the commonalities and differences among different works, and finally provide a discussion regarding the challenges and future directions for this field.

Change detection based on heterogeneous images, such as optical images and synthetic aperture radar images, is a challenging problem because of their huge appearance differences. To combat this problem, we propose an unsupervised change detection method that contains only a convolutional autoencoder (CAE) for feature extraction and the commonality autoencoder for commonalities exploration. The CAE can eliminate a large part of redundancies in two heterogeneous images and obtain more consistent feature representations. The proposed commonality autoencoder has the ability to discover common features of ground objects between two heterogeneous images by transforming one heterogeneous image representation into another. The unchanged regions with the same ground objects share much more common features than the changed regions. Therefore, the number of common features can indicate changed regions and unchanged regions, and then a difference map can be calculated. At last, the change detection result is generated by applying a segmentation algorithm to the difference map. In our method, the network parameters of the commonality autoencoder are learned by the relevance of unchanged regions instead of the labels. Our experimental results on five real data sets demonstrate the promising performance of the proposed framework compared with several existing approaches.

Evolutionary algorithms and swarm intelligence algorithms have been widely used for constrained optimization problems for decades and numerous techniques for constraint handling have been proposed. The${\varepsilon }$-constrained method is a very effective one. In the literature, the${\varepsilon }$value was usually controlled via an exponential function, which is not competent for solving certain types of constrained optimization problems, e.g., whose global optima are located near the boundary of the feasible and infeasible regions. To solve this problem, this article proposes a new adaptive${\varepsilon }$control method and incorporate it into a basic differential evolution (DE) algorithm: (DE/rand/1/exp). Based on the information of constraint violation in the current population, the adaptive method controls the value of${\varepsilon }$through a simple heuristic rule. Compared with the traditional exponential function-based control methods, the proposed adaptive method can prevent the algorithm from being trapped into local optima while retaining the obtained near-optimal candidate solutions in the infeasible region for generating promising searching paths. Besides, we set the crossover rate (CR) as a more reasonable value for DE/rand/1/exp, which can enhance the efficiency significantly. The well-known 2006 IEEE Congress on Evolutionary Computation (CEC 2006) competition on real-parameter single-objective constrained optimization benchmark is adopted to evaluate the effectiveness of the proposed adaptive${\varepsilon }$-constrained DE. Fifteen constrained engineering optimization problems are collected from the literature to test the proposed algorithm. Moreover, the adaptive${\varepsilon }$control method is extended to an adaptive algorithm to solve the benchmark problems from CEC 2017. The comparison results confirm the superiority of the proposed method.

Deep neural networks (DNNs) have been frequently and widely applied for intelligent systems such as object detection, natural language understanding and speech recognition. Given a specific problem, we always aim to construct the most suitable DNN to solve it, which requires choosing the most appropriate model architecture and seeking the best model parameters values. However, most existing works focus on model parameters learning under the assumption that the model architecture can be manually specified as per prior knowledge and/or trial-and-error experimentation. To overcome this problem, evolutionary algorithms (EAs) have been widely used to design model architectures automatically. Further, EAs have been used for neural network optimization for more than 30 years. Therefore, in this paper, we review the evolutionary neural architecture search (ENAS) from the view of the advanced techniques. We hope this work can provide a comprehensive understanding of EAs' roles for the readers and focus themselves on ENAS.

Though it is well known that the performance of deep neural networks (DNNs) degrades under certain light conditions, there exists no study on the threats of light beams emitted from some physical source as adversarial attacker on DNNs in a real-world scenario. In this work, we show by simply using a laser beam that DNNs are easily fooled. To this end, we propose a novel attack method called Adversarial Laser Beam (AdvLB), which enables manipulation of laser beam’s physical parameters to perform adversarial attack. Experiments demonstrate the effectiveness of our proposed approach in both digital- and physical-settings. We further empirically analyze the evaluation results and reveal that the proposed laser beam attack may lead to some interesting prediction errors of the state-of-the-art DNNs. We envisage that the proposed AdvLB method enriches the current family of adversarial attacks and builds the foundation for future robustness studies for light.

Human can easily recognize visual objects with lost information: even losing most details with only contour reserved, e.g. cartoon. However, in terms of visual perception of Deep Neural Networks (DNNs), the ability for recognizing abstract objects (visual objects with lost information) is still a challenge. In this work, we investigate this issue from an adversarial viewpoint: will the performance of DNNs decrease even for the images only losing a little information? Towards this end, we propose a novel adversarial attack, named AdvDrop, which crafts adversarial examples by dropping existing information of images. Previously, most adversarial attacks add extra disturbing information on clean images explicitly. Opposite to previous works, our proposed work explores the adversarial robustness of DNN models in a novel perspective by dropping imperceptible de-tails to craft adversarial examples. We demonstrate the effectiveness of AdvDrop by extensive experiments, and show that this new type of adversarial examples is more difficult to be defended by current defense systems.

Deep reinforcement learning (DRL) has demonstrated promising performance to learn effective heuristics to solve complex combinatorial optimisation problems via policy networks. However, traditional reinforcement learning (RL) suffers from insufficient exploration, which often results in pre-convergence to poor policies and many challenges the performance of DRL. To prevent this, we propose an Entropy Regularised Reinforcement Learning (ERRL) method that supports exploration by providing more stochastic policies, improving optimisation. The ERRL method incorporates an entropy term, defined over the policy network's outputs, into the loss function of the policy network. Hence, policy exploration can be explicitly advocated subjected to a balance to maximise the reward. As a result, the risk of pre-convergence to inferior policies can be reduced. We implement the ERRL method based on two existing DRL algorithms. We have compared the performances of our implementations with the two DRL algorithms along with several state-of-the-art heuristic-based non-RL approaches for three categories of routing problems, i.e., travelling salesman problem (TSP), capacitated vehicle routing problem (CVRP) and multiple routing with fixed fleet problems (MRPFF). Experimental results show that the proposed method can find better and faster solutions in most test cases than the state-of-the-art algorithms.

With the emergence of crowdshipping and sharing economy, vehicle routing problem with occasional drivers (VRPOD) has been recently proposed to involve occasional drivers with private vehicles for the delivery of goods. In this article, we present a generalized variant of VRPOD, namely, the vehicle routing problem with heterogeneous capacity, time window, and occasional driver (VRPHTO), by taking the capacity heterogeneity and time window of vehicles into consideration. Furthermore, to meet the requirement in today's cloud computing service, wherein multiple optimization tasks may need to be solved at the same time, we propose a novel evolutionary multitasking algorithm (EMA) to optimize multiple VRPHTOs simultaneously with a single population. Finally, 56 new VRPHTO instances are generated based on the existing common vehicle routing benchmarks. Comprehensive empirical studies are conducted to illustrate the benefits of the new VRPHTOs and to verify the efficacy of the proposed EMA for multitasking against a state-of-art single task evolutionary solver. The obtained results showed that the employment of occasional drivers could significantly reduce the routing cost, and the proposed EMA is not only able to solve multiple VRPHTOs simultaneously but also can achieve enhanced optimization performance via the knowledge transfer between tasks along the evolutionary search process.

Automated construction of deep neural networks (DNNs) has become a research hot spot nowadays because DNN’s performance is heavily influenced by its architecture and parameters, which are highly task-dependent, but it is notoriously difficult to find the most appropriate DNN in terms of architecture and parameters to best solve a given task. In this work, we provide an insight into the automated DNN construction process by formulating it into a multilevel multiobjective large-scale optimization problem with constraints, where the nonconvex, nondifferentiable, and black-box nature of this problem make evolutionary algorithms (EAs) to stand out as a promising solver. Then, we give a systematical review of existing evolutionary DNN construction techniques from different aspects of this optimization problem and analyze the pros and cons of using EA-based methods in each aspect. This work aims to help DNN researchers to better understand why, where, and how to utilize EAs for automated DNN construction and meanwhile, help EA researchers to better understand the task of automated DNN construction so that they may focus more on EA-favored optimization scenarios to devise more effective techniques.

Similar trajectory search is a crucial task that facilitates many downstream spatial data analytic applications. Despite its importance, many of the current literature focus solely on the trajectory’s spatial similarity while neglecting the temporal information. Additionally, the few papers that use both the spatial and temporal features based their approach on a traditional point-to-point comparison. These methods model the importance of the spatial and temporal aspect of the data with only a single, pre-defined balancing factor for all trajectories, even though the relative spatial and temporal balance can change from trajectory to trajectory. In this article, we propose the first spatio-temporal, deep-representation-learning-based approach to similar trajectory search. Experiments show that utilizing both features offers significant improvements over existing point-to-point comparison and deep-representation-learning approach. We also show that our deep neural network approach is faster and performs more consistently compared to the point-to-point comparison approaches.

With the prevalence of the Internet, online reviews have become a valuable information resource for people. However, the authenticity of online reviews remains a concern, and deceptive reviews have become one of the most urgent network security problems to be solved. Review spams will mislead users into making suboptimal choices and inflict their trust in online reviews. Most existing research manually extracted features and labeled training samples, which are usually complicated and time-consuming. This paper focuses primarily on a neglected emerging domain - movie review, and develops a novel unsupervised spam detection model with an attention mechanism. By extracting the statistical features of reviews, it is revealed that users will express their sentiments on different aspects of movies in reviews. An attention mechanism is introduced in the review embedding, and the conditional generative adversarial network is exploited to learn users’ review style for different genres of movies. The proposed model is evaluated on movie reviews crawled from Douban, a Chinese online community where people could express their feelings about movies. The experimental results demonstrate the superior performance of the proposed approach.

Deep neural networks (DNNs), characterized by sophisticated architectures capable of learning a hierarchy of feature representations, have achieved remarkable successes in various applications. Learning DNN's parameters is a crucial but challenging task that is commonly resolved by using gradient-based backpropagation (BP) methods. However, BP-based methods suffer from severe initialization sensitivity and proneness to getting trapped into inferior local optima. To address these issues, we propose a DNN learning framework that hybridizes CC-based optimization with BP-based gradient descent, called BPCC, and implement it by devising a computationally efficient CC-based optimization technique dedicated to DNN parameter learning. In BPCC, BP will intermittently execute for multiple training epochs. Whenever the execution of BP in a training epoch cannot sufficiently decrease the training objective function value, CC will kick in to execute by using the parameter values derived by BP as the starting point. The best parameter values obtained by CC will act as the starting point of BP in its next training epoch. In CC-based optimization, the overall parameter learning task is decomposed into many subtasks of learning a small portion of parameters. These subtasks are individually addressed in a cooperative manner. In this article, we treat neurons as basic decomposition units. Furthermore, to reduce the computational cost, we devise a maturity-based subtask selection strategy to selectively solve some subtasks of higher priority. Experimental results demonstrate the superiority of the proposed method over common-practice DNN parameter learning techniques.

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples. Existing works have mostly focused on either digital adversarial examples created via small and imperceptible perturbations, or physical-world adversarial examples created with large and less realistic distortions that are easily identified by human observers. In this paper, we propose a novel approach, called Adversarial Camouflage (\emph{AdvCam}), to craft and camouflage physical-world adversarial examples into natural styles that appear legitimate to human observers. Specifically, \emph{AdvCam} transfers large adversarial perturbations into customized styles, which are then “hidden” on-target object or off-target background. Experimental evaluation shows that, in both digital and physical-world scenarios, adversarial examples crafted by \emph{AdvCam} are well camouflaged and highly stealthy, while remaining effective in fooling state-of-the-art DNN image classifiers. Hence, \emph{AdvCam} is a flexible approach that can help craft stealthy attacks to evaluate the robustness of DNNs.

Predictive analytics in Mobile Edge Computing (MEC) based Internet of Things (IoT) is becoming a high demand in many real-world applications. A prediction problem in an MEC-based IoT environment typically corresponds to a collection of tasks with each task solved in a specific MEC environment based on the data accumulated locally, which can be regarded as a Multi-task Learning (MTL) problem. However, the heterogeneity of the data (non-IIDness) accumulated across different MEC environments challenges the application of general MTL techniques in such a setting. Federated MTL (FMTL) has recently emerged as an attempt to address this issue. Besides FMTL, there exists another powerful but under-exploited distributed machine learning technique, called Network Lasso (NL), which is inherently related to FMTL but has its own unique features. In this paper, we made an in-depth evaluation and comparison of these two techniques on three distinct IoT datasets representing real-world application scenarios. Experimental results revealed that NL outperformed FMTL in MEC-based IoT environments in terms of both accuracy and computational efficiency.

Precise segmentation of organs and tumors plays a crucial role in clinical applications. It is a challenging task due to the irregular shapes and various sizes of organs and tumors as well as the significant class imbalance between the anatomy of interest (AOI) and the background region. In addition, in most situation tumors and normal organs often overlap in medical images, but current approaches fail to delineate both tumors and organs accurately. To tackle such challenges, we propose a position-guided deformable UNet, namely PGD-UNet, which exploits the spatial deformation capabilities of deformable convolution to deal with the geometric transformation of both organs and tumors. Position information is explicitly encoded into the network to enhance the capabilities of deformation. Meanwhile, we introduce a new pooling module to preserve position information lost in conventional max-pooling operation. Besides, due to unclear boundaries between different structures as well as the subjectivity of annotations, labels are not necessarily accurate for medical image segmentation tasks. It may cause the overfitting of the trained network due to label noise. To address this issue, we formulate a novel loss function to suppress the influence of potential label noise on the training process. Our method was evaluated on two challenging segmentation tasks and achieved very promising segmentation accuracy in both tasks.

Conventional DNN training paradigms typically rely on one training set and one validation set, obtained by partitioning an annotated dataset available for the purpose of training, namely gross training set, in a certain way. The training set is used for training the model while the validation set is used to estimate the generalization performance of the trained model as the training proceeds to avoid over-fitting. There exist two major issues in this training paradigm. Firstly, the validation set may hardly guarantee an unbiased estimate of the generalization performance due to potential mismatching with the test data. Secondly, training a DNN corresponds to solve a complex optimization problem, which is prone to getting trapped into inferior local optima and thus leads to the undesired training result. To address these issues, we propose a novel DNN training framework. It generates multiple pairs of training and validation sets from the gross training set via random splitting, trains a DNN model of a pre-specified network structure on each pair while making the useful knowledge (e.g., promising network parameters) obtained from one model training process to be transferred to other model training processes via multi-task optimization (i.e., a recently emerging optimization paradigm), and outputs the best one, among all trained models, which has the overall best performance across the validation sets from all pairs. The knowledge transfer mechanism featured in this new framework can not only enhance training effectiveness by helping the model training process to escape from local optima but also improve on generalization performance via implicit regularization imposed on one model training process from other model training processes. We implement the proposed framework, parallelize the implementation on a GPU cluster, and apply it to train several widely used DNN models. Experimental results on several classification datasets of different nature demonstrate the superiority of the proposed framework over the conventional training paradigm.

Evolutionary multitask optimization (EMTO) is a newly emerging research area in the field of evolutionary computation. It investigates how to solve multiple optimization problems (tasks) at the same time via evolutionary algorithms (EAs) to improve on the performance of solving each task independently, assuming if some component tasks are related then the useful knowledge (e.g., promising candidate solutions) acquired during the process of solving one task may assist in (and also benefit from) solving the other tasks. In EMTO, task relatedness is typically unknown in advance and needs to be captured via EA's population. Since the population of an EA can only cover a subregion of the solution space and keeps evolving during the search, thus captured task relatedness is local and dynamic. The multifactorial EA (MFEA) is one of the most representative EMTO techniques, inspired by the bio-cultural model of multifactorial inheritance, which transmits both biological and cultural traits from the parents to the offspring. MFEA has succeeded in solving various multitask optimization (MTO) problems. However, the intensity of knowledge transfer in MFEA is determined via its algorithmic configuration without considering the degree of task relatedness, which may prevent the effective sharing and utilization of the useful knowledge acquired in related tasks. To address this issue, we propose a self-regulated EMTO (SREMTO) algorithm to automatically adapt the intensity of cross-task knowledge transfer to different and varying degrees of relatedness between different tasks as the search proceeds so that the useful knowledge in common for solving related tasks can be captured, shared, and utilized to a great extent. We compare SREMTO with MFEA and its variants as well as the single-task optimization counterpart of SREMTO on two MTO test suites, which demonstrates the superiority of SREMTO.

Image change detection detects the regions of change in multiple images of the same scene taken at different times, which plays a crucial role in many applications. The two most popular image change detection techniques are as follows: pixel-based methods heavily rely on accurate image coregistration while object-based approaches can tolerate coregistration errors to some extent but are sensitive to image segmentation or classification errors. To address these issues, we propose an unsupervised image change detection approach based on a novel bipartite differential neural network (BDNN). The BDNN is a deep neural network with two input ends, which can extract the holistic features from the unchanged regions in the two input images, where two learnable change disguise maps (CDMs) are used to disguise the changed regions in the two input images, respectively, and thus demarcate the unchanged regions therein. The network parameters and CDMs will be learned by optimizing an objective function, which combines a loss function defined as the likelihood of the given input image pair over all possible input image pairs and two constraints imposed on CDMs. Compared with the pixel-based and object-based techniques, the BDNN is less sensitive to inaccurate image coregistration and does not involve image segmentation or classification. In fact, it can even skip over coregistration if the degree of transformation (due to the different view angles and/or positions of the camera) between the two input images is not that large. We compare the proposed approach with several state-of-the-art image change detection methods on various homogeneous and heterogeneous image pairs with and without coregistration. The results demonstrate the superiority of the proposed approach.

We investigate the matrix function optimization under the orthonormal constraint on the matrix variable. By introducing an index-notation-arrangement-based chain rule (I-Chain rule), we obtain the gradient of the cost function and propose a revisited orthonormal-constraint-based projected gradient method to locate a minimum of an objective/cost function of matrix variables iteratively subject to orthonormal constraint. To guarantee the convergence the proposed method, existing schemes require the gradient can be represented by the multiplication of a symmetrical matrix and the matrix variable itself. This condition has been relaxed in this paper. New techniques are proposed to establish the convergence property of the iterative algorithm. Simulation results show the effectiveness of our framework. This paper allows more extensive applications of matrix function optimization problems in science and engineering.

Evolutionary multi-task optimization (EMTO) is a newly emerging research area which studies on how to solve multiple optimization problems simultaneously using evolutionary algorithms (EAs) so that useful knowledge (e.g. promising candidate solutions) obtained when solving one task can be transferred and reused to facilitate solving some other tasks. In EMTO, how to effectively transfer and reuse knowledge among multiple tasks is an important subject of study. Among existing EMTO techniques, DE-based EMTO algorithms deserve special attention because DE, as one of the most popular EAs, has achieved remarkable feats for solving challenging (single-task) optimization problems in the past although DE-based EMTO is not yet extensively studied. In this paper, we propose a general multitasking DE (MTDE) framework which aims to help more clearly understand the focuses and differences of existing and potential works on DE-based EMTO. We implement this framework via three commonly-used DE search schemes and perform a thorough empirical study on two DE-specific knowledge reuse strategies which are based upon base and differential vectors, respectively. Experiments on 18 MTO demonstrate that the base vector based knowledge reuse strategy outperforms the differential vector based one across all three tested DE search schemes.

Multi-task optimization (MTO) is a newly emerging research area in the field of optimization, studying on how to solve multiple optimization problems at the same time so that the processes of solving different but relevant problems could help each other via knowledge transfer to improve the overall performance of solving all problems. Evolutionary MTO (EMTO) employs evolutionary algorithms as the optimizer and treats the candidate solutions that perform commonly well on multiple tasks as the transferable knowledge between these tasks. In this work, we propose a multitasking multi-swarm optimization (MTMSO) algorithm which extends a popular dynamic multi-swarm optimization (DMS-PSO) algorithm into the multitasking scenario. In MTMSO, the whole swarm is randomly partitioned into multiple swarms (i.e., task groups) with each being responsible for solving a specific task, and each swarm is further partitioned into multiple sub-swarms. Within each task group, optimization is performed as per the mechanism of DMS-PSO for solving a specific task. Cross-task knowledge transfer is realized via probabilistic crossover of the personal bests of the particles from different task groups. Both task groups and each group's sub-swarms are periodically reformed to maintain search diversity. An adaptive local search process, featuring dynamic allocation of the computational resource for each task, is incorporated in the final stage of optimization to improve the quality of the best solution found for each task. The proposed MTMSO algorithm is compared with a single-task conventional PSO and a popular EMTO algorithm on two test suites composing of 9 simple and 10 complex single-objective MTO problems, respectively, which demonstrates its superiority.

Evolutionary multi-task optimization (EMTO) studies on how to simultaneously solve multiple optimization problems, so-called component problems, via evolutionary algorithms, which has drawn much attention in the field of evolutionary computation. Knowledge transfer across multiple optimization problems (being solved) is the key to make EMTO to outperform traditional optimization paradigms. In this work, we propose a simple and effective knowledge transfer strategy which utilizes the best solution found so far for one problem to assist in solving the other problems during the optimization process. This strategy is based on random replacement. It does not introduce extra computational cost in terms of objective function evaluations for solving each component problem. However, it helps to improve optimization effectiveness and efficiency, compared to solving each component problem in a standalone way. This light-weight knowledge transfer strategy is implemented via differential evolution within a multi-population based EMTO paradigm, leading to a differential evolutionary multi-task optimization (DEMTO) algorithm. Experiments are conducted on the CEC'2017 competition test bed to compare the proposed DEMTO algorithm with five state-of-the-art EMTO algorithms, which demonstrate the superiority of DEMTO.

We propose a distributed machine-learning architecture to predict trustworthiness of sensor services in Mobile Edge Computing (MEC) based Internet of Things (IoT) services, which aligns well with the goals of MEC and requirements of modern IoT systems. The proposed machine-learning architecture models training a distributed trust prediction model over a topology of MEC-environments as a Network Lasso problem, which allows simultaneous clustering and optimization on large-scale networked-graphs. We then attempt to solve it using Alternate Direction Method of Multipliers (ADMM) in a way that makes it suitable for MEC-based IoT systems. We present analytical and simulation results to show the validity and efficiency of the proposed solution.

Actions always refer to complex vision variations in a long-range redundant video sequence. Instead of focusing on limited range sequence, i.e. convolution on adjacent frames, in this paper, we proposed an action recognition approach with bootstrapping based long-range temporal context attention. Specifically, due to vision variations of the local region across frames, we target at capturing temporal context by proposing the Temporal Pixels based Parallel-head Attention (TPPA) block. In TPPA, we apply the self-attention mechanism between local regions at the same position across temporal frames to capture the interaction impacts. Meanwhile, to deal with video redundancy and capture long-range context, the TPPA is extended to the Random Frames based Bootstrapping Attention (RFBA) framework. While the bootstrapping sampling frames have the same distribution of the whole video sequence, the RFBA not only captures longer temporal context with only a few sampling frames but also has comprehensive representation through multiple sampling. Furthermore, we also try to apply this temporal context attention to image-based action recognition, by transforming the image into "pseudo video" with the spatial shift. Finally, we conduct extensive experiments and empirical evaluations on two most popular datasets:UCF101 for videos andStanford40 for images. In particular, our approach achieves top-1 accuracy of $91.7%$ in UCF101 and mAP of $90.9%$ in Stanford40.

A fundamental challenge in real-time labelling of activity data is user burden. The Experience Sampling Method (ESM) is widely used to obtain such labels for sensor data. However, in an in-situ deployment, it is not feasible to expect users to precisely label the start and end time of each event or activity. For this reason, time-point based experience sampling (without an actual start and end time) is prevalent. We present a framework that applies multi-instance and semi-supervised learning techniques to perform to predict user annotations from multiple mobile sensor data streams. Our proposed framework estimates users' annotations in ESM-based studies progressively, via an interactive pipeline of co-training and active learning. We evaluate our work using data collected from an in-the-wild data collection.

Evolutionary multitasking (EMT) is an emerging research topic in the field of evolutionary computation. In contrast to the traditional single-task evolutionary search, EMT conducts evolutionary search on multiple tasks simultaneously. It aims to improve convergence characteristics across multiple optimization problems at once by seamlessly transferring knowledge among them. Due to the efficacy of EMT, it has attracted lots of research attentions and several EMT algorithms have been proposed in the literature. However, existing EMT algorithms are usually based on a common mode of knowledge transfer in the form of implicit genetic transfer through chromosomal crossover. This mode cannot make use of multiple biases embedded in different evolutionary search operators, which could give better search performance when properly harnessed. Keeping this in mind, this paper proposes an EMT algorithm with explicit genetic transfer across tasks, namely EMT via autoencoding, which allows the incorporation of multiple search mechanisms with different biases in the EMT paradigm. To confirm the efficacy of the proposed EMT algorithm with explicit autoencoding, comprehensive empirical studies have been conducted on both the single- and multi-objective multitask optimization problems.

Service composition is a key enabling technology in service-oriented computing for developing versatile applications by integrating various existing interoperable services. Although actively studied, most existing works on service composition neglect the existence of complementarity among candidate services within a service class, so-called internal complementarity. In fact, complementary candidate services within a service class can be composed to form a composite candidate service which may yield better service utility than that provided by any existing candidate service within that service class. This work focuses on web service composition where internal complementarity is more likely to happen. Specifically, we aim at addressing the problem of QoS-aware web service selection with internal complementarity (WSS-IC). We first transform this problem into a multi-choice multi-dimensional knapsack problem (MMKP) and prove such a transformation per se has non-polynomial time complexity in the worse case.Then, we perform complexity analysis to demonstrate that existing approaches to MMKPs are not computationally feasible to resolve QoS-aware WSS-IC. This fact motivates us to propose an iteratively improving framework for deriving the solution iteration by iteration while taking into account both solution structure and QoS constraints. At each iteration, the current solution gets improved by solving a disjunctively constrained knapsack problem. To verify the effectiveness of the proposed framework, two heuristic approaches are implemented under this framework. Experimental results demonstrate that our approaches outperform the compared methods in terms of both solution quality and computation time.

Ensemble learning is a powerful machine learning paradigm which leverages a collection of diverse base learners to achieve better prediction performance than that could be achieved by any individual base learner. This work proposes an evolutionary feature subspaces generation based ensemble learning framework, which formulates the tasks of searching for the most suitable feature subspace for each base learner into a multi-task optimization problem and solve it via an evolutionary multi-task optimizer. Multiple such problems which correspond to different base learners are solved simultaneously via an evolutionary multi-task feature selection algorithm such that solving one problem may help solve some other problems via implicit knowledge transfer. The quality of thus generated feature subspaces is supposed to outperform those obtained by individually seeking the optimal feature subspace for each base learner. We implement the proposed framework by using SVM, KNN, and decision tree as the base learners, proposing a multi-task binary particle swarm optimization algorithm for evolutionary multi-task feature selection, and utilizing the major voting scheme to combine the outputs of the base learners. Experiments on several UCI datasets demonstrate the effectiveness of the proposed method.

Energy consumption prediction typically corresponds to a multivariate time series prediction task where different channels in the multivariate time series represent energy consumption data and various auxiliary data related to energy consumption such as environmental factors. It is non-trivial to resolve this task, which requires finding the most appropriate prediction model and the most useful features (extracted from the raw data) to be used by the model. This work proposes an evolutionary multi-objective ensemble learning (EMOEL) technique which uses extreme learning machines (ELMs) as base predictors due to its highly recognized efficacy. EMOEL employs evolutionary multi-objective optimization to search for the optimal parameters of the model as well as the optimal features fed into the model subjected to two conflicting criteria, i.e., accuracy and diversity. It leads to a Pareto front composed of non-dominated optimal solutions where each solution depicts the number of hidden neurons in the ELM, the selected channels in the multivariate time series, the selected feature extraction methods and the selected time windows applied to the selected channels. The optimal solutions in the Pareto front stand for different end-to-end prediction models which may lead to different prediction results. To boost ultimate prediction accuracy, the models with respect to these optimal solutions are linearly combined with combination coefficients being optimized via an evolutionary algorithm. We evaluate the proposed method in comparison to some existing prediction techniques on an Australian University based dataset, which demonstrates the superiority of the proposed method.

In this paper, we focus on simultaneous inference of transportation modes and human activities in daily life via modelling and inference from multivariate time series data, which are streamed from off-the-shelf mobile sensors (e.g. embedded in smartphones) in real-world dynamic environments. The transportation mode will be inferred from the structured hierarchical contexts associated with human activities. Through our mobile context recognition system, an accurate and robust solution can be obtained to infer transportation mode, human activity and their associated contexts (e.g. whether the user is in moving or stationary environment) simultaneously. There are many challenges in analysing and modelling human mobility patterns within urban areas due to the ever-changing environments of mobile users. For instance, a user could stay at a particular location and then travel to various destinations depending on the tasks they carry within a day. Consequently, there is a need to reduce the reliance on location-based sensors (e.g. GPS), since they consume a significant amount of energy on smart devices, for the purpose of intelligent mobile sensing (i.e. automatic inference of transportation mode, human activity and associated contexts). Nevertheless, our system is capable of outperforming the simplistic approach that only considers independent classifications of multiple context label sets on data streamed from low-energy sensors.

Recently, door access control with Internet of Things has become increasingly popular in the field of security. However, conventional approaches such as video-based or biological information-based cannot satisfy the requirements of personal privacy protection in the modern society. Hence, a wireless signal-based technique which does not need users to carry any devices, called device-free have been introduced in recent years to detect and identify persons. In this paper, we present BLEDoorGuard, a wireless, invisible, and robust door access system which leverages received signal strength indicator from Bluetooth low energy (BLE) beacons to recognize a person who accesses a door. We evaluated BLEDoorGuard in two real-world scenarios: the first is an office with a key lock, and the second is a meeting room with swipe card access. We exploit the characteristics of use of BLE for person identification and propose a two-step algorithm with multiple classifiers. We demonstrate that BLEDoorGuard is capable of identifying the actual user during door access with an accuracy of 69% and 62% among groups of 6 and 10 people, respectively.

We propose an unsupervised deep convolutional coupling network for change detection based on two heterogeneous images acquired by optical sensors and radars on different dates. Most existing change detection methods are based on homogeneous images. Due to the complementary properties of optical and radar sensors, there is an increasing interest in change detection based on heterogeneous images. The proposed network is symmetric with each side consisting of one convolutional layer and several coupling layers. The two input images connected with the two sides of the network, respectively, are transformed into a feature space where their feature representations become more consistent. In this feature space, the different map is calculated, which then leads to the ultimate detection map by applying a thresholding algorithm. The network parameters are learned by optimizing a coupling function. The learning process is unsupervised, which is different from most existing change detection methods based on heterogeneous images. Experimental results on both homogenous and heterogeneous images demonstrate the promising performance of the proposed network compared with several existing approaches.

We propose a novel dynamic metaheuristic optimization approach to compose an optimal set of IaaS service requests to align with an IaaS provider's long-term economic expectation. This approach is designed for the context that the IaaS provisioning subjects to resource and QoS constraints. In addition, the IaaS service requests have the features of dynamic resource and QoS requirements and variable arrival times. A new economic model is proposed to evaluate the similarity between the provider's long-term economic expectation and a composition of service requests. The evaluation incorporates the factors of dynamic pricing and operation cost modeling of the service requests. An innovative hybrid genetic algorithm is proposed that incorporates the economic inter-dependency among the requests as a heuristic operator and performs repair operations in local solutions to meet the resource and QoS constraints. The proposed approach generates dynamic global solutions by updating the heuristic operator at regular intervals with the runtime behavior data of an existing service composition. Experimental results preliminarily prove the feasibility of the proposed approach.

Brain storm optimization (BSO) is a newly emerging family of swarm intelligence techniques inspired by the human's creative problem-solving process, which has achieved successes in many applications. BSO is characterized by its unique process of grouping a population of ideas and carrying out brainstorming based on the grouped ideas to search for optima generation by generation. Although the original BSO is a sequential algorithm based on the central processing unit (CPU), its major algorithmic modules are highly suitable for parallelization. Nowadays, modern graphic processing units (GPUs) have become widely affordable, which empower personal computers to undertake massively parallel computing tasks. Therefore, this work investigates a GPU-based implementation of BSO using NVIDIA's CUDA technology, aiming to accelerate BSO's computation speed while maintaining its optimization accuracy. Experimental results on 30 CEC2014 single-objective real-parameter optimization benchmark problems demonstrate the remarkable speedups of the proposed GPU-based parallel BSO compared to the original CPU-based sequential BSO across varying problems and population sizes.

We investigate the use of subjective metrics in social media to evaluate cloud service performance in the market. We first examine the subjective factors that drive cloud consumers to/from purchasing cloud services. These include the ability to achieve greater scalability, security concerns, etc. according to several industry surveys. We then analyse the correlation between the consumers' perception on those factors and the cloud market revenue growth. This paper identifies the unique subjective metrics that are indicative of cloud service performance from the market perspective. The cloud consumers' perception is sourced from several particular social media using sentiment analysis techniques. We focus on consumers' perception on a leading cloud provider that holds the majority of the cloud market share. We find that subjective metrics are empirically proved to be applicable in evaluating the performance of cloud services in the market.

In numerous industrial applications where safety, efficiency, and reliability are among primary concerns, condition-based maintenance (CBM) is often the most effective and reliable maintenance policy. Prognostics, as one of the key enablers of CBM, involves the core task of estimating the remaining useful life (RUL) of the system. Neural networks-based approaches have produced promising results on RUL estimation, although their performances are influenced by handcrafted features and manually specified parameters. In this paper, we propose a multiobjective deep belief networks ensemble (MODBNE) method. MODBNE employs a multiobjective evolutionary algorithm integrated with the traditional DBN training technique to evolve multiple DBNs simultaneously subject to accuracy and diversity as two conflicting objectives. The eventually evolved DBNs are combined to establish an ensemble model used for RUL estimation, where combination weights are optimized via a single-objective differential evolution algorithm using a task-oriented objective function. We evaluate the proposed method on several prognostic benchmarking data sets and also compare it with some existing approaches. Experimental results demonstrate the superiority of our proposed method.

The Steiner tree problem in graphs (STPG) is a well known NP-hard combinatorial problem with various applications in transport, computational biology, network and VLSI design. Exact methods have been developed to solve this problem to proven optimality, however the exponential nature of these algorithms mean that they become intractable with large-scale instances of the problem. Because of this phenomenon, there has been considerable research into using metaheuristics to obtain good quality solutions in a reasonable time. This paper presents a hybrid local search technique which is an extension of techniques from the literature with an added random jump operator which prevents the algorithm from becoming stuck in local minima. It is compared against greedy local search, the hybrid local search technique it extends and two metaheuristic techniques from the current literature and is shown to outperform them in nearly all cases.

With the deep penetration of the Internet and mobile devices, privacy preservation in the local setting has become increasingly relevant. The local setting refers to the scenario where a user is willing to share his/her information only if it has been properly sanitized before leaving his/her own device. Moreover, a user may hold only a single data element to share, instead of a database. Despite its ubiquitousness, the above constraints make the local setting substantially more challenging than the traditional centralized or distributed settings. In this paper, we initiate the study of private spatial data aggregation in the local setting, which finds its way in many real-world applications, such as Waze and Google Maps. In response to users' varied privacy requirements that are natural in the local setting, we propose a new privacy model called personalized local differential privacy (PLDP) that allows to achieve desirable utility while still providing rigorous privacy guarantees. We design an efficient personalized count estimation protocol as a building block for achieving PLDP and give theoretical analysis of its utility, privacy and complexity. We then present a novel framework that allows an untrusted server to accurately learn the user distribution over a spatial domain while satisfying PLDP for each user. This is mainly achieved by designing a novel user group clustering algorithm tailored to our problem. We confirm the effectiveness and efficiency of our framework through extensive experiments on multiple real benchmark datasets.

Given a child's and a couple's facial photos, tri-subject kinship verification aims to determine the existence of blood relation between the child and the couple. Different from existing methods which model the kinship inheritance process among three persons in separate stages and only use simple features, this work establishes a simple model inspired by genetics to measure tri-subject kinship similarity in one step. Meanwhile, high-dimensional features are incorporated into this simple model to seek for better performance. Experiment results demonstrate the effectiveness of our approach.

Road detection from images is a challenging task in computer vision. Previous methods are not robust, because their features and classifiers cannot adapt to different circumstances. To overcome this problem, we propose to apply unsupervised feature learning for road detection. Specifically, we develop an improved encoding function and add a feature selection process to obtain robust and discriminative road features. Besides, a road segmentation algorithm is proposed to extract road regions from the learned feature maps, in which a tree structure is established to represent the hierarchical relations of various regions segmented by multiple thresholds, and a two-loop optimization is then employed to select the most stable regions as road areas. Experimental results on several challenging datasets justify the effectiveness of our method.

If output signals of artificial neural network classifiers are interpreted per node as class label predictors then partial knowledge encoded by the network during the learning procedure can be exploited in order to reassign which output node should represent each class label so that learning speed and final classification accuracy are improved. Our method for computing these reassignments is based on the maximum average correlation between actual node outputs and target labels over a small labeled validation dataset. Node Label Matching is an ancillary method for both supervised and unsupervised learning in artificial neural networks and we demonstrate its integration with Contrastive Divergence pre-training in Restricted Boltzmann Machines and Back Propagation fine-tuning in Deep Belief Networks. We introduce the Segmented Density Random Binary dataset and present empirical results of Node Label Matching on both our synthetic data and a subset of the MNIST benchmark.

Auto-encoder is a popular representation learning technique which can capture the generative model of data via a encoding and decoding procedure typically driven by reconstruction errors in an unsupervised way. In this paper, we propose a semi-supervised manifold learning based auto-encoder (named semAE). semAE is based on a regularized auto-encoder framework which leverages semi-supervised manifold learning to impose regularization based on the encoded representation. Our proposed approach suits more practical scenarios in which a small number of labeled data are available in addition to a large number of unlabeled data. Experiments are conducted on several well-known benchmarking datasets to validate the efficacy of semAE from the aspects of both representation and classification. The comparisons to state-of-the-art representation learning methods on classification performance in semi-supervised settings demonstrate the superiority of our approach.

In this paper, Extreme Learning Machine (ELM) is demonstrated to be a powerful tool for electricity consumption prediction based on its competitive prediction accuracy and superior computational speed compared to Support Vector Machine (SVM). Moreover, ELM is utilized to investigate the potentials of using auxiliary information such as electricity-related factors and environmental factors to augment the prediction accuracy obtained by purely using the electricity consumption factors. Furthermore, we formulate a combinatorial optimization problem of seeking an optimal subset of auxiliary factors and their corresponding optimal window sizes using the most suitable ELM structure, and propose a Discrete Dynamic Multi-Swarm Particle Swarm Optimization (DDMS-PSO) to address this problem. Experimental studies on a real-world building dataset demonstrate that electricity-related factors improve accuracy while environmental factors further boost accuracy. By using DDMSPSO, we find a subset of electricity-related and environmental factors, their respective window sizes, and the number of hidden neurons in ELM which leads to the best prediction accuracy.

Multi-activity recognition in the urban environment is a challenging task. This is largely attributed to the influence of urban dynamics, the variety of the label sets, and the heterogeneous nature of sensor data that arrive irregularly and at different rates. One of the first tasks in multi-activity recognition is temporal segmentation. A common temporal segmentation method is the sliding window approach with a fixed window size, which is widely used for single activity recognition. In order to recognise multiple activities from heterogeneous sensor streams, we propose a new time windowing technique that can optimally extract segments with multiple activity labels. The mixture of activity labels causes the impurity in the corresponding temporal segment. Hence, larger window size imposes higher impurity in temporal segments while increasing class separability. In addition, the combination of labels from multiple activity label sets (i.e. number of unique multi-activity) may decrease as impurity increases. Naturally, these factors will affect the performance of classification task. In our proposed technique, the optimal window size is found by gaining the balance between minimising impurity and maximising class separability in temporal segments. As a result, it accelerates the learning process for recognising multiple activities (such as higher level and atomic human activities under different environment contexts) in comparison to laborious tasks of sensitivity analysis. The evaluation was validated by experiments on a real-world dataset for recognising multiple human activities in a smart environment.

Differential evolution (DE) is a powerful population-based stochastic optimization algorithm. Although its efficacy has been witnessed in various applications, the performance of DE is usually challenged when the computational budget is decreased and/or the search landscape's complexity is increased. To address these issues, we propose a new local ensemble surrogate assisted crowding DE (LES-CDE) algorithm, which consists of multiple local surrogate models built upon the historical search information accumulated in diverse overlapped local regions of the search space. In LES-CDE, an ensemble of several adjacent local surrogates is utilized to guide the creation of promising trial vectors. To maintain the local nature of each surrogate model, LES-CDE uses the replacement scheme of crowding DE (CDE) to update the population which also serves as model landmarks. We test LES-CDE under varying parameters and compare them with CDE on 15 numerical test problems taken from CEC 2015 single-objective real-parameter optimization testbed. Results from our experiments demonstrate the superiority of LES-CDE over CDE in a statistically significant manner.

Cooperative Co-evolutionary (CC) techniques have demonstrated the promising performance in dealing with large-scale optimization problems. However, in many applications, their performance may drop due to the presence of imbalanced contributions to the objective function value from different subsets of decision variables. To remedy this drawback, Contribution-Based Cooperative Co-evolutionary (CBCC) algorithms have been proposed. They have presented significant improvements over traditional CC techniques when the decomposition is accurate and the imbalance level is very high. However, in real-world scenarios, we might not have the knowledge about the ideal decomposition and actual imbalance level of a problem to be solved. Therefore, this study aims at analysing the performance of existing CBCC techniques in more realistic settings, i.e., when the decomposition error is unavoidable and the imbalance level is low or moderate. Our in-depth analysis reveals that even in these situations, CBCC algorithms are superior alternatives to traditional CC techniques. We also observe that the variations of CBCC techniques may lead to the significantly different performance. Thus, we recommend practitioners to carefully choose a competent variant of CBCC which best suits their particular applications.

Cloud service composition problem (CSCP) is usually long-term based in practice. A logical request is to maximize end users' long-term benefit. Thus, the overall long-term QoS properties of the composite service should be optimized and the users' requirements during the period should be satisfied. However, the benefit-maximization has not been considered under the background of long-term based CSCP in existing research yet. To fill this gap, in this paper, a new formulation LCSCP is proposed to define the long-term based CSCP as an optimization problem. Then, for the sake of efficiency, three meta-heuristic approaches (i.e, Genetic Algorithm, Simulated Annealing and Tabu Search) are studied. Comprehensive experiments are designed and conducted to test their various aspects of performance on different test sets with different workflows. Experimental results provide a basic perspective of how these three widely adopted meta-heuristic frameworks work on this new problem, which can be baseline work for further research.

We propose a novel composition framework for an Infrastructure-as-a-Service (IaaS) provider that selects the optimal set of long-term service requests to maximize its profit. Existing solutions consider an IaaS provider's economic benefits at the time of service composition and ignore the dynamic nature of the consumer requests in a long-term period. The proposed framework deploys a new multivariate HMM and ARIMA model to predict different patterns of resource utilization and Quality of Service fluctuation tolerance levels of existing service consumers. The dynamic nature of new consumer requests with no history is modelled using a new community based heuristic approach. The predicted long-term service requests are optimized using Integer Linear Programming to find a proper configuration that maximizes the profit of an IaaS provider. Experimental results prove the feasibility of the proposed approach.

Hyperspectral image classification is a challenging problem. Among existing approaches to addressing this problem, the active learning (AL) and semisupervised learning (SSL) techniques have attracted much attention in recent years. AL usually involves a labor-intensive human-labeling process while SSL, although avoiding human labeling by assigning pseudolabels to unlabeled data, may introduce incorrect pseudolabels and thus deteriorate classification performance. To overcome these drawbacks, a novel approach named collaborative active and semisupervised learning (CASSL) is proposed in this paper. CASSL combines AL and SSL to invoke a collaborative labeling process by both human experts and classifiers. Specifically, an AL-based pseudolabel verification procedure is performed for gradually improving the pseudolabeling accuracy to facilitate SSL. Meanwhile, only those unlabeled data with low pseudolabeling confidence in SSL will become the query candidates in AL. We evaluate the performance of CASSL on three hyperspectral data sets and compare it with that of two state-of-the-art hyperspectral image classification methods. Experimental results reveal the superiority of CASSL.

This work provides an in-depth investigation of the effects of population initialization on Differential Evolution (DE) for dealing with large scale optimization problems. Firstly, we conduct a statistical parameter sensitive analysis to study the effects of DE's control parameters on its performance of solving large scale problems. This study reveals the optimal parameter configurations which can lead to the statistically superior performance over the CEC-2013 large-scale test problems. Thus identified optimal parameter configurations interestingly favour much larger population sizes while agreeing with the other parameter settings compared to the most commonly employed parameter configuration. Based on one of the identified optimal configurations and the most commonly used configuration, which only differ in the population size, we investigate the influence of various population initialization techniques on DE's performance. This study indicates that initialization plays a more crucial role in DE with a smaller population size. However, this observation might be the result of insufficient convergence due to the use of a large population size under the limited computational budget, which deserve more investigations.

Although various population initialization techniques have been employed in evolutionary algorithms (EAs), there lacks a comprehensive survey on this research topic. To fill this gap and attract more attentions from EA researchers to this crucial yet less explored area, we conduct a systematic review of the existing population initialization techniques. Specifically, we categorize initialization techniques from three exclusive perspectives, i.e., randomness, compositionality and generality. Characteristics of the techniques belonging to each category are carefully analysed to further lead to several sub-categories. We also discuss several open issues related to this research topic, which demands further in-depth investigations.

Opposition-based learning (OBL) and cooperative co-evolution (CC) have demonstrated promising performance when dealing with large-scale global optimization (LSGO) problems. In this work, we propose a novel framework for hybridizing these two techniques, and investigate the performance of simple implementations of this new framework using the most recent LSGO benchmarking test suite. The obtained results verify the effectiveness of our proposed OBL-CC framework. Moreover, some advanced statistical analyses reveal that the proposed hybridization significantly outperforms its component methods in terms of the quality of finally obtained solutions.

Differential evolution (DE), as a very powerful population-based stochastic optimizer, is one of the most active research topics in the field of evolutionary computation. Self-adaptive differential evolution (SaDE) is a well-known DE variant, which aims to relieve the practical difficulty faced by DE in selecting among many candidates the most effective search strategy and its associated parameters. SaDE operates with multiple candidate strategies and gradually adapts the employed strategy and its accompanying parameter setting via learning the preceding behavior of already applied strategies and their associated parameter settings. Although highly effective, SaDE concentrates more on exploration than exploitation. To enhance SaDE's exploitation capability while maintaining its exploration power, we incorporate local search chains into SaDE following two different paradigms (Lamarckian and Baldwinian) that differ in the ways of utilizing local search results in SaDE. Our experiments are conducted on the CEC-2014 real-parameter single-objective optimization testbed. The statistical comparison results demonstrate that SaDE with Baldwinian local search chains, armed with suitable parameter settings, can significantly outperform original SaDE as well as classic DE at any tested problem dimensionality.

Feature selection aims to select a subset of relevant features from typically a large number of original features, which is a difficult task due to the large search space. Particle swarm optimisation (PSO) is a powerful search technique, but there are some limitations on using the standard PSO for feature selection. This paper proposes a new PSO based feature selection approach, which introduces an external archive to store promising solutions obtained during the search process. The solutions in the archive serve as potential leaders (i.e. global best, gbest) to guide the swarm to search for an optimal feature subset with the lowest classification error rate and a smaller number of features. The proposed approach has two specific methods, PSOArR and PSOArRWS, where PSOArR randomly selects gbest from the archive and PSOArRWS uses the roulette wheel selection to select gbest considering both the classification error rate and also considering the number of selected features. Experiments on twelve benchmark datasets show that both PSOArR and PSOArRWS can successfully select a smaller number of features and achieve similar or better classification performance than using all features. PSOArR and PSOArRWS outperform a PSO based algorithm without using an archive and two traditional feature selection methods. The performance of PSOArR and PSOArRWS are similar to each other.

Digital photo management is becoming indispensable for the explosively growing family photo albums due to the rapid popularization of digital cameras and mobile phone cameras. An effective photo management system could accurately and efficiently group all faces of the same person into a small number of clusters. In this paper, we present a novel photo grouping method based on spectral theory. The key idea is to utilize prior information of family photo albums to improve the performance. First, an individual can only appear once in one photo, which works as the similarity constraint in our graph construction. Second, an individual cannot show more times than the number of photos in each album. That is, the size of a cluster for an individual is at most the number of photos in an album. We consider this constraint as a Minimum Cost Flow (MCF) linear network optimization problem and therefore propose a constrained K-Means for data clustering after graph embedding. Two metrics, i.e., accuracy (AC) and normalized mutual information metric (NMI), are used to evaluate the clustering performance. Extensive experimental results demonstrate the effectiveness of the proposed method.

The convex hull has been extensively studied in computational geometry and its applications have spread over an impressive number of fields. How to find the convex hull is an important and challenging problem. Although many algorithms had been proposed for that, most of them can only tackle the problem in two or three dimensions and the biggest issue is that those algorithms rely on the samples' coordinates to find the convex hull. In this paper, we propose an approximation algorithm named FVDM, which only utilizes the information of the samples' distance matrix to find the convex hull. Experiments demonstrate that FVDM can effectively identify the vertices of the convex hull.

Several population initialization methods for evolutionary algorithms (EAs) have been proposed previously. This paper categorizes the most well-known initialization methods and studies the effect of them on large scale global optimization problems. Experimental results indicate that the optimization of large scale problems using EAs is more sensitive to the initial population than optimizing lower dimensional problems. Statistical analysis of results show that basic random number generators, which are the most commonly used method for population initialization in EAs, lead to the inferior performance. Furthermore, our study shows, regardless of the size of the initial population, choosing a proper initialization method is vital for solving large scale problems.

Differential evolution (DE) is one of the most powerful continuous optimizers in the field of evolutionary computation. This work systematically benchmarks a classic DE algorithm (DE/rand/1/bin) on the CEC-2013 single-objective continuous optimization testbed. We report, for each test function at different problem dimensionality, the best achieved performance among a wide range of potentially effective parameter settings. It reflects the intrinsic optimization capability of DE/rand/1/bin on this testbed and can serve as a baseline for performance comparison in future research using this testbed. Furthermore, we conduct parameter sensitivity analysis using advanced non-parametric statistical tests to discover statistically significantly superior parameter settings. This analysis provides a statistically reliable rule of thumb for choosing the parameters of DE/rand/1/bin to solve unseen problems. Moreover, we report the performance of DE/rand/1/bin using one superior parameter setting advocated by parameter sensitivity analysis.

Self-adaptive differential evolution (SaDE) is a wellknown DE variant, which has received considerable attention since it was developed. SaDE gradually adapts its trial vector generation strategy and the accompanying parameter setting via learning the preceding performance of multiple candidate strategies and their associated parameter settings. This work systematically investigates SaDE on the CEC-2013 real-parameter single-objective optimization testbed. Parameter sensitivity analysis is carried out by using advanced statistical hypothesis testing methods, aiming to detect statistically significantly superior parameter settings. This analysis reveals that SaDE is actually less sensitive to the parameter choice since quite a number of parameter settings can lead to the statistically significantly better performance than the other settings. Based on this finding, we report SaDE's performance using one of the parameter settings advocated by sensitivity analysis and statistically compare this performance with that of a widely used classic DE (DE/rand/1/bin). The comparison results significantly favor SaDE.

Compared to conventional activity recognition methods using feature extraction followed by classification, the Genetic Programming (GP) based classification applied to raw sensor data can avoid the time-consuming and knowledge-dependent feature extraction procedure. However, the traditional GP-based classifier using accuracy as fitness function is sensitive to the choice of threshold values. Furthermore, sensor data of the same activity might demonstrate remarkable distinction when the signal is collected in the changing environment, which will lead to inconsistency between training and testing data and consequently degrade the generalization power of the trained classifier. Moreover, the GP-based classifier cannot well distinguish less separable activities in the presence of multiple activities. Our work aims to address these issues by improving the GP-based classifier via: (1) using the area under the receiver operating characteristic curve (AUC) as fitness function, (2) using an online local time series normalization procedure to pre-smooth undesirable features, and (3) using a binary tree based classification framework to force GP to learn key discriminating features that can better distinguish less separable activities. We test the proposed method on a sensor data set collected from a smartphone, consisting of four common human activities, sitting, standing, walking and running. The proposed GP-based classifier achieves the outstanding performance on recognizing each of four activities in terms of both high true positive and low false alarm rates, which much improves over the traditional GP-based classifier and several of its variants.

We propose a novel model to mine and derive class-specific codebook for categorical object detection and classification. In particular, the codebook is built from a pool of heterogeneous local descriptors using an effective feature selection scheme. The resulting class-specific codebook strengthens the class discriminability by learning the most discriminative part codewords constructed from their preferable local descriptors. The advantage of our class-specific codebook comes from two aspects. 1). As we collect a variety of heterogeneous descriptors during the learning of local codebook, each target object class can always be represented by its most preferable descriptors. Moreover, even each part codeword can also find its suitable descriptors. 2). The feature selection process further picks out the most discriminative object parts that separate the target object class from background and other classes. Experimental results on several widely used datasets show that benefits from our class-specific object codebook which fuses complementary visual cues remarkably improve the detection and classification performance for both rigid and non-rigid articulated objects.

Modern GPUs enable widely affordable personal computers to carry out massively parallel computation tasks. NVIDIA's CUDA technology provides a wieldy parallel computing platform. Many state-of-the-art algorithms arising from different fields have been redesigned based on CUDA to achieve computational speedup. Differential evolution (DE), as a very promising evolutionary algorithm, is highly suitable for parallelization owing to its data-parallel algorithmic structure. However, most existing CUDA-based DE implementations suffer from excessive low-throughput memory access and less efficient device utilization. This work presents an improved CUDA-based DE to optimize memory and device utilization: several logically-related kernels are combined into one composite kernel to reduce global memory access; kernel execution configuration parameters are automatically determined to maximize device occupancy; streams are employed to enable concurrent kernel execution to maximize device utilization. Experimental results on several numerical problems demonstrate superior computational time efficiency of the proposed method over two recent CUDA-based DE and the sequential DE across varying problem dimensions and algorithmic population sizes.

A region-based unsupervised segmentation and classification algorithm for polarimetric synthetic aperture radar (SAR) imagery that incorporates region growing and a Markov random field edge strength model is designed and implemented. This algorithm is an extension of the successful Iterative Region Growing with Semantics (IRGS) segmentation and classification algorithm, which was designed for amplitude only SAR imagery, to polarimetric data. Polarimetric IRGS (PolarIRGS) extends IRGS by incorporating a polarimetric feature model based on the Wishart distribution and modifying key steps such as initialization, edge strength computation, and the region growing criterion. Like IRGS, PolarIRGS oversegments an image into regions and employs iterative region growing to reduce the size of the solution search space. The incorporation of an edge penalty in the spatial context model improves segmentation performance by preserving segment boundaries that traditional spatial models will smooth over. Evaluation of PolarIRGS with Flevoland fully polarimetric data shows that it improves upon two other recently published techniques in terms of classification accuracy.

Harmony search (HS), as an emerging metaheuristic technique mimicking the improvisation behavior of musicians, has demonstrated strong efficacy in solving various numerical and real-world optimization problems. This work presents a harmony search with differential mutation based pitch adjustment (HSDM) algorithm, which improves the original pitch adjustment operator of HS using the self-referential differential mutation scheme that features differential evolution - another celebrated metaheuristic algorithm. In HSDM, the differential mutation based pitch adjustment can dynamically adapt the properties of the landscapes being explored at different searching stages. Meanwhile, the pitch adjustment operator's execution probability is allowed to vary randomly between 0 and 1, which can maintain both wild and fine exploitation throughout the searching course. HSDM has been evaluated and compared to the original HS and two recent HS variants using 16 numerical test problems of various searching landscape complexities at 10 and 30 dimensions. HSDM almost always demonstrates superiority on all test problems.

Multivariate image segmentation is a challenging task, influenced by large intraclass variation that reduces class distinguishability as well as increased feature space sparseness and solution space complexity that impose computational cost and degrade algorithmic robustness. To deal with these problems, a Markov random field (MRF) based multivariate segmentation algorithm called "multivariate iterative region growing using semantics" (MIRGS) is presented. In MIRGS, the impact of intraclass variation and computational cost are reduced using the MRF spatial context model incorporated with adaptive edge penalty and applied to regions. Semantic region growing starting from watershed over-segmentation and performed alternatively with segmentation gradually reduces the solution space size, which improves segmentation effectiveness. As a multivariate iterative algorithm, MIRGS is highly sensitive to initial conditions. To suppress initialization sensitivity, it employs a region-level k -means (RKM) based initialization method, which consistently provides accurate initial conditions at low computational cost. Experiments show the superiority of RKM relative to two commonly used initialization methods. Segmentation tests on a variety of synthetic and natural multivariate images demonstrate that MIRGS consistently outperforms three other published algorithms.

Differential evolution (DE) is an efficient and powerful population-based stochastic search technique for solving optimization problems over continuous space, which has been widely applied in many scientific and engineering fields. However, the success of DE in solving a specific problem crucially depends on appropriately choosing trial vector generation strategies and their associated control parameter values. Employing a trial-and-error scheme to search for the most suitable strategy and its associated parameter settings requires high computational costs. Moreover, at different stages of evolution, different strategies coupled with different parameter settings may be required in order to achieve the best performance. In this paper, we propose a self-adaptive DE (SaDE) algorithm, in which both trial vector generation strategies and their associated control parameter values are gradually self-adapted by learning from their previous experiences in generating promising solutions. Consequently, a more suitable generation strategy along with its parameter settings can be determined adaptively to match different phases of the search process/evolution. The performance of the SaDE algorithm is extensively evaluated (using codes available from P. N. Suganthan) on a suite of 26 bound-constrained numerical optimization problems and compares favorably with the conventional DE and several state-of-the-art parameter adaptive DE variants.

In this paper, our recently developed Self-adaptive Differential Evolution algorithm (SaDE) is extended to solve numerical optimization problems with multiple conflicting objectives. The performance of the proposed MOSaDE algorithm is evaluated on a suit of 19 benchmark problems provided for the CEC2007 special session (http://www.ntu.edu.sg/home/epnsugan/) on Performance Assessment of Multi-Objective Optimization Algorithms.

In this paper, we propose an extension of Self-adaptive Differential Evolution algorithm (SaDE) to solve optimization problems with constraints. In comparison with the original SaDE algorithm, the replacement criterion was modified for handling constraints. The performance of the proposed method is reported on the set of 24 benchmark problems provided by CEC2006 special session on constrained real parameter optimization.

This paper presents a palmprint recognition system with palmprint images collected by a high-resolution color scanner. The scanned RGB image of the palmprint is pre-processed and the region of interest (ROI) of the palm is determined by the finger gap locations. Three sets of features extracted from the ROI image by the 2D Gabor filter using the palmprint phase orientation code (PPOC) represent texture information of the palm in the form of a real component, an imaginary component and an orientation component, respectively. The recognition is performed by applying the enhanced linear discriminant analysis (EDLDA) coupled with the nearest neighbor classifier on these three feature sets, respectively, and the decisions are combined via the majority voting scheme to yield the ultimate recognition. Experiments on our collected palmprint image database show promising recognition rate of 99.6% with a low False Acceptance Rate (FAR) of 0.02%

This paper presents a variant of particle swarm optimizers (PSOs) that we call the comprehensive learning particle swarm optimizer (CLPSO), which uses a novel learning strategy whereby all other particles' historical best information is used to update a particle's velocity. This strategy enables the diversity of the swarm to be preserved to discourage premature convergence. Experiments were conducted (using codes available from http://www.ntu.edu.sg/home/epnsugan) on multimodal test functions such as Rosenbrock, Griewank, Rastrigin, Ackley, and Schwefel and composition functions both with and without coordinate rotation. The results demonstrate good performance of the CLPSO in solving multimodal problems when compared with eight other recent variants of the PSO.

In this paper, we propose a novel self-adaptive differential evolution algorithm (SaDE), where the choice of learning strategy and the two control parameters F and CR are not required to be pre-specified. During evolution, the suitable learning strategy and parameter settings are gradually self-adapted according to the learning experience. The performance of the SaDE is reported on the set of 25 benchmark functions provided by CEC2005 special session on real parameter optimization.