Abstract Multi-camera depth estimation has gained significant attention in autonomous driving due to its importance in perceiving complex environments. However, extending monocular self-supervised methods to multi-camera setups introduces unique challenges that existing techniques often fail to address. In this paper, we propose STViT+ , a novel Transformer-based framework for self-supervised multi-camera depth estimation. Our key contributions include: 1) the Spatial-Temporal Transformer (STTrans) , which integrates local spatial connectivity and global context to capture enriched spatial-temporal cross-view correlations, resulting in more accurate 3D geometry reconstruction; 2) the Spatial-Temporal Photometric Consistency Correction (STPCC) strategy that mitigates the impact of varying illumination, ensuring brightness consistency across frames during photometric loss calculation; 3) the Adversarial Geometry Regularization (AGR) module, which employs Generative Adversarial Networks to impose spatial constraints by using unpaired depth maps, enhancing performance under adverse conditions such as rain and nighttime driving. Extensive evaluations on large-scale autonomous driving datasets, including Nuscenes and DDAD, confirm that STViT+ sets a new benchmark for multi-camera depth estimation.

To defend the inference attacks and mitigate the sensitive information leakages in Federated Learning (FL), client-level Differentially Private FL (DPFL) is the de-facto standard for privacy protection by clipping local updates and adding random noise. However, existing DPFL methods tend to make a sharp loss landscape and have poor weight perturbation robustness, resulting in severe performance degradation. To alleviate these issues, we propose a novel DPFL algorithm named DP-FedSAM, which leverages gradient perturbation to mitigate the negative impact of DP. Specifically, DP-FedSAM integrates Sharpness Aware Minimization (SAM) optimizer to generate local flatness models with improved stability and weight perturbation robustness, which results in the small norm of local updates and robustness to DP noise, thereby improving the performance. To further reduce the magnitude of random noise while achieving better performance, we propose DP-FedSAM-$\operatorname{top}_{k}$topk by adopting the local update sparsification technique. From the theoretical perspective, we present the convergence analysis to investigate how our algorithms mitigate the performance degradation induced by DP. Meanwhile, we give rigorous privacy guarantees with Rényi DP, the sensitivity analysis of local updates, and generalization analysis. At last, we empirically confirm that our algorithms achieve state-of-the-art (SOTA) performance compared with existing SOTA baselines in DPFL.

For multichannel synthetic aperture radar (SAR) ground moving target identification (GMTI) in heterogeneous and strong clutter backgrounds, it is challenging to accurately detect slow and weak targets when relying solely on the magnitude of clutter-suppression residuals, due to the significant clutter residuals and signal-to-noise losses. To address this, a detector that leverages both the magnitude and phase information in multichannel SAR-GMTI clutter suppression is proposed. For aM-channel SAR system, the detection test is formulated as the product of the residual magnitude fromM-channel clutter suppression and a phase factor derived from the interferometric phase between two residuals from the first and lastM- 1 channels. This phase factor captures the dissimilarity from the clutter, enabling the suppression of strong clutter residuals and improving the signal-to-clutter-plus-noise ratio (SCNR). Using the product clutter model, a constant false alarm ratio detection framework is designed. The receiver operator characteristic metrics, obtained from simulations and real-data experiments, validate the proposed method’s superiority over the state-of-the-art techniques, and the detection sensitivity on the clutter heterogeneity, correlation coefficients between pairs of clutter-suppression residuals, and target parameters is analyzed for practicality. In the X-band airborne radar GMTI experiments, the minimum discernible input SCNR of -6 dB for target radial velocity of 4 m/s and 0 dB for 2 m/s demonstrate the effectiveness in detecting the dim targets within strong clutter.

In ground-based L-band radar sea clutter, Bragg scattering caused by short gravity waves on the sea surface frequently exhibits azimuthal dependence, with higher order Bragg-scattering spectra clearly visible alongside the first-order spectrum. Recent measurements from an L-band airborne moving target detection (MTD) radar, operating in side-looking mode with HH polarization at medium incident angles (30°–60°), near the Zhoushan Fishing Ground in Ningbo, China, also reveal azimuth-dependent and multipeak characteristics, that pose challenges for target detection within the endo-clutter region. To better understand the clutter characteristics in L-band airborne MTD radar, this article investigates the modeling of sea clutter Doppler spectra under medium incident angles ranging from 30° to 60°. Using the small slope approximation (SSA) incorporating a time-varying rough sea surface with spikes, a systemic expression for the Doppler spectrum, accounting for the sea clutter space-time coupling, is derived. Specifically, the Doppler spectrum related to the sea surface is expressed as an azimuth-dependent underlying spectrum weighted by the antenna beam (i.e., the spatial spectrum), while the spectrum due to spikes is represented as a convolution of the spatial spectrum with an azimuth-independent underlying spectrum. Each individual spectrum is characterized with Gaussian profiles, forming the basis of a comprehensive spectrum model that can successfully capture an azimuth-independent peak and approximately 2–7 or more azimuth-dependent peaks in the real-world sea-clutter Doppler spectra. Note that the proposed model is directed against sea echoes with azimuth-dependent scattering properties from the main lobe of the two-way antenna pattern, and thus, suitably characterizes the corresponding sea-clutter Doppler features.

Space teleoperation significantly extends human reach in space missions. However, traditional approaches are constrained by factors, such as the reliance on accurate dynamic models and the risk of operator fatigue during prolonged tasks. Additionally, while data-driven intelligent approaches reduce the need for prior knowledge, they have yet to adequately address the time delay issues inherent in these systems. To overcome these challenges, we introduce the belief state actor-critic (BSAC) method, the first deep reinforcement learning approach tailored for space teleoperation capture tasks within a bilateral control framework. We first establish a generalized agent-based architecture for space teleoperation, shifting decision-making from human operators to autonomous agents. Following a comprehensive analysis of the time delay challenges, we propose the BSAC algorithm, which integrates state augmentation and belief state techniques to mitigate the effects of delays in teleoperated Markov decision processes. Extensive experiments are conducted on the MuJoCo simulation platform, modeling a real hardware system across various scenarios. The learned policies are then successfully transferred and validated in a real-world setup, demonstrating the effectiveness and robustness of BSAC. In summary, our results support the feasibility of agent-based frameworks capable of overcoming time delay challenges in space teleoperation.

3D-aware GANs successfully solve the problem of 3D-consistency generation and furthermore provide a 3D shape of the generated object. However, the application of the volume renderer disturbs the disentanglement of the latent space, which makes it difficult to manipulate 3D-aware GANs and lowers the image quality of style-based generators. In this work, we devise a dual-mapping framework to make the generated images of pretrained 2D StyleGAN consistent in 3D space. We utilize a tri-plane representation to estimate the 3D shape of the generated object and two mapping networks to bridge the latent space of StyleGAN and the 3D tri-plane space. Our method does not alter the parameters of the pretrained generator, which means the interpretability of latent space is preserved for various image manipulations. Experiments show that our method lifts the 3D awareness of pretrained 2D StyleGAN to 3D-aware GANs and outperforms the 3D-aware GANs in controllability and image quality.

Multi-camera depth estimation has recently garnered significant attention due to its substantial practical implications in the realm of autonomous driving. In this paper, we delve into the task of self-supervised multi-camera depth estimation and propose an innovative framework, STViT, featuring several noteworthy enhancements: 1) we propose a Spatial-Temporal Transformer to comprehensively exploit both local connectivity and the global context of image features, meanwhile learning enriched spatial-temporal cross-view correlations to recover 3D geometry. 2) to alleviate the severe effect of adverse conditions, e.g., rainy weather and nighttime driving, we introduce a GAN-based Adversarial Geometry Regularization Module (AGR) to further constrain the depth estimation with unpaired normal-condition depth maps and prevent the model from being incorrectly trained. Experiments on challenging autonomous driving datasets Nuscenes and DDAD show that our method achieves state-of-the-art performance.

The occurrence of frame drops due to issues such as corrupted communications or malfunctioning sensors presents a significant challenge to an agent’s decision-making, especially in remote control scenarios. Classical reinforcement learning (RL) usually assumes a continuous data stream without frame drops and relies heavily on online interactions, which is time-consuming, resource-intensive, and often impractical in certain scenarios. Consequently, the performance of RL may deteriorate significantly in face of non-negligible frame drops. To tackle this challenge caused by frame dropping, We propose Conditional Diffusion Model for Decision-Making under Random Frame Dropping (D3D), an offline algorithm that can effectively enhance performance robustness in frame dropping scenarios. D3D addresses this issue through a two-phase approach: 1) During the policy generation phase, D3D adopts a return-conditional diffusion model for decision making rather than the temporal difference learning, whose policy is derived using offline datasets of return-labeled trajectories without information loss. 2) When frame dropping occurs during evaluation, D3D seamlessly substitutes the missing state with its corresponding prediction in the horizon made by the diffusion model. Extensive experiments are conducted on MuJoCo and Adroit tasks to validate D3D’s robustness and efficiency. The results demonstrate that D3D consistently outperforms state-of-the-art RL algorithms, especially excelling on tasks featuring severe drop rates.

Federated learning (FL) can train collaboratively with several mobile terminals (MTs), which faces critical challenges in communication, resource, and privacy. Existing privacy-preserving methods usually adopt instance-level differential privacy (DP), which provides a rigorous privacy guarantee but with several bottlenecks: performance degradation, transmission overhead, and resource constraints. Therefore, we propose Fed-LTP, an efficient and privacy-enhanced FL framework withLotteryTicketHypothesis (LTH) and zero-concentrated DP(zCDP). It generates a pruned global model on the server side and conducts sparse-to-sparse training from scratch with zCDP on the client side. On the server side, two pruning schemes are proposed: (i) the weight-based pruning (LTH) determines the pruned global model structure; (ii) the iterative pruning further shrinks the size of the pruned model. Meanwhile, the performance of Fed-LTP is boosted via model validation based on the Laplace mechanism. On the client side, we use sparse-to-sparse training to solve the resource-constraints issue and provide tighter privacy analysis to reduce the privacy budget. We evaluate the effectiveness of Fed-LTP on several real-world datasets in both independent and identically distributed (IID) and non-IID settings. The results confirm the superiority of Fed-LTP over state-of-the-art (SOTA) methods in communication, computation, and memory efficiencies while realizing a better utility-privacy trade-off.

A key challenge of continual reinforcement learning (CRL) in dynamic environments is to promptly adapt the reinforcement learning (RL) agent's behavior as the environment changes over its lifetime while minimizing the catastrophic forgetting of the learned information. To address this challenge, in this article, we propose DaCoRL, that is, dynamics-adaptive continual RL. DaCoRL learns a context-conditioned policy using progressive contextualization, which incrementally clusters a stream of stationary tasks in the dynamic environment into a series of contexts and opts for an expandable multihead neural network to approximate the policy. Specifically, we define a set of tasks with similar dynamics as an environmental context and formalize context inference as a procedure of online Bayesian infinite Gaussian mixture clustering on environment features, resorting to online Bayesian inference to infer the posterior distribution over contexts. Under the assumption of a Chinese restaurant process (CRP) prior, this technique can accurately classify the current task as a previously seen context or instantiate a new context as needed without relying on any external indicator to signal environmental changes in advance. Furthermore, we employ an expandable multihead neural network whose output layer is synchronously expanded with the newly instantiated context and a knowledge distillation regularization term for retaining the performance on learned tasks. As a general framework that can be coupled with various deep RL algorithms, DaCoRL features consistent superiority over existing methods in terms of stability, overall performance, and generalization ability, as verified by extensive experiments on several robot navigation and MuJoCo locomotion tasks.

Safety comes first in many real-world applications involving autonomous agents. Despite a large number of reinforcement learning (RL) methods focusing on safety-critical tasks, there is still a lack of high-quality evaluation of those algorithms that adheres to safety constraints at each decision step under complex and unknown dynamics. In this paper, we revisit prior work in this scope from the perspective of state-wise safe RL and categorize them as projection-based, recovery-based, and optimization-based approaches, respectively. Furthermore, we propose Unrolling Safety Layer (USL), a joint method that combines safety optimization and safety projection. This novel technique explicitly enforces hard constraints via the deep unrolling architecture and enjoys structural advantages in navigating the trade-off between reward improvement and constraint satisfaction. To facilitate further research in this area, we reproduce related algorithms in a unified pipeline and incorporate them into SafeRL-Kit, a toolkit that provides off-the-shelf interfaces and evaluation utilities for safety-critical tasks. We then perform a comparative study of the involved algorithms on six benchmarks ranging from robotic control to autonomous driving. The empirical results provide an insight into their applicability and robustness in learning zero-cost-return policies without task-dependent handcrafting. The project page is available at https://sites.google.com/view/saferlkit.

To mitigate the privacy leakages and communication burdens of Federated Learning (FL), decentralized FL (DFL) discards the central server and each client only communicates with its neighbors in a decentralized communication network. However, existing DFL suffers from high inconsistency among local clients, which results in severe distribution shift and inferior performance compared with centralized FL (CFL), especially on heterogeneous data or sparse communication topologies. To alleviate this issue, we propose two DFL algorithms named DFedSAM and DFedSAM-MGS to improve the performance of DFL. Specifically, DFedSAM leverages gradient perturbation to generate local flat models via Sharpness Aware Minimization (SAM), which searches for models with uniformly low loss values. DFedSAM-MGS further boosts DFedSAM by adopting Multiple Gossip Steps (MGS) for better model consistency, which accelerates the aggregation of local flat models and better balances communication complexity and generalization. Theoretically, we present improved convergence rates $\small \mathcal{O}\big(\frac{1}{\sqrt{KT}}+\frac{1}{T}+\frac{1}{K^{1/2}T^{3/2}(1-\lambda)^2}\big)$ and $\small \mathcal{O}\big(\frac{1}{\sqrt{KT}}+\frac{1}{T}+\frac{\lambda^Q+1}{K^{1/2}T^{3/2}(1-\lambda^Q)^2}\big)$ in non-convex setting for DFedSAM and DFedSAM-MGS, respectively, where $1-\lambda$ is the spectral gap of gossip matrix and $Q$ is the number of MGS. Empirically, our methods can achieve competitive performance compared with CFL methods and outperform existing DFL methods.

Data sparsity is a common issue for most recommender systems and can severely degrade the usefulness of a system. One of the most successful solutions to this problem has been cross-domain recommender systems. These frameworks supplement the sparse data of the target domain with knowledge transferred from a source domain rich with data that is in some way related. However, there are three challenges that, if overcome, could significantly improve the quality and accuracy of cross-domain recommendation: 1) ensuring latent feature spaces of the users and items are both maximally matched; 2) taking consideration of user-item relationship and their interaction in modelling user preference; 3) enabling a two-way cross-domain recommendation that both the source and the target domains benefit from a knowledge exchange. Hence, in this paper, we propose a novel deep neural network called Dual Adversarial network for Cross-Domain Recommendation (DA-CDR). By training the shared encoders with a domain discriminator via dual adversarial learning, the latent feature spaces for both the users and items are maximally matched between the source and target domains. The domain-specific encoders are applied with an orthogonal constraint to ensure that any domain-specific features are properly extracted and work as supplement to the shared features. Allowing the two domains to collaboratively benefit from each other results in better recommendations for both domains. Extensive experiments with real-world datasets on six tasks demonstrate that DA-CDR significantly outperforms seven state-of-the-art baselines in terms of recommendation accuracy.

To solve the user data sparsity problem, which is the main issue in generating user preference prediction, cross-domain recommender systems transfer knowledge from one source domain with dense data to assist recommendation tasks in the target domain with sparse data. However, data are usually sparsely scattered in multiple possible source domains, and in each domain (source/target) the data may be heterogeneous, thus it is difficult for existing cross-domain recommender systems to find one source domain with dense data from multiple domains. In this way, they fail to deal with data sparsity problems in the target domain and cannot provide an accurate recommendation. In this article, we propose a novel multidomain recommender system (called HMRec) to deal with two challenging issues: 1) how to exploit valuable information from multiple source domains when no single source domain is sufficient and 2) how to ensure positive transfer from heterogeneous data in source domains with different feature spaces. In HMRec, domain-shared and domain-specific features are extracted to enable the knowledge transfer between multiple heterogeneous source and target domains. To ensure positive transfer, the domain-shared subspaces from multiple domains are maximally matched by a multiclass domain discriminator in an adversarial learning process. The recommendation in the target domain is completed by a matrix factorization module with aligned latent features from both the user and the item side. Extensive experiments on four cross-domain recommendation tasks with real-world datasets demonstrate that HMRec can effectively transfer knowledge from multiple heterogeneous domains collaboratively to increase the rating prediction accuracy in the target domain and significantly outperforms six state-of-the-art non-transfer or cross-domain baselines.

The powerful learning ability of deep neural networks enables reinforcement learning (RL) agents to learn competent control policies directly from continuous environments. In theory, to achieve stable performance, neural networks assume identically and independently distributed (i.i.d.) inputs, which unfortunately does not hold in the general RL paradigm where the training data are temporally correlated and nonstationary. This issue may lead to the phenomenon of "catastrophic interference" and the collapse in performance. In this article, we present interference-aware deep Q-learning (IQ) to mitigate catastrophic interference in single-task deep RL. Specifically, we resort to online clustering to achieve on-the-fly context division, together with a multihead network and a knowledge distillation regularization term for preserving the policy of learned contexts. Built upon deep Q networks (DQNs), IQ consistently boosts the stability and performance when compared to existing methods, verified with extensive experiments on classic control and Atari tasks. The code is publicly available at https://github.com/ Sweety-dm/Interference-aware-Deep-Q-learning.

In the unsupervised open set domain adaptation (UOSDA), the target domain contains unknown classes that are not observed in the source domain. Researchers in this area aim to train a classifier to accurately: 1) recognize unknown target data (data with unknown classes) and 2) classify other target data. To achieve this aim, a previous study has proven an upper bound of the target-domain risk, and the open set difference, as an important term in the upper bound, is used to measure the risk on unknown target data. By minimizing the upper bound, a shallow classifier can be trained to achieve the aim. However, if the classifier is very flexible [e.g., deep neural networks (DNNs)], the open set difference will converge to a negative value when minimizing the upper bound, which causes an issue where most target data are recognized as unknown data. To address this issue, we propose a new upper bound of target-domain risk for UOSDA, which includes four terms: source-domain risk,$\epsilon $-open set difference ($\Delta _\epsilon $), distributional discrepancy between domains, and a constant. Compared with the open set difference,$\Delta _\epsilon $is more robust against the issue when it is being minimized, and thus we are able to use very flexible classifiers (i.e., DNNs). Then, we propose a new principle-guided deep UOSDA method that trains DNNs via minimizing the new upper bound. Specifically, source-domain risk and$\Delta _\epsilon $are minimized by gradient descent, and the distributional discrepancy is minimized via a novel open set conditional adversarial training strategy. Finally, compared with the existing shallow and deep UOSDA methods, our method shows the state-of-the-art performance on several benchmark datasets, including digit recognition [modified National Institute of Standards and Technology database (MNIST), the Street View House Number (SVHN), U.S. Postal Service (USPS)], object recognition (Office-31, Office-Home), and face recognition [pose, illumination, and expression (PIE)].

One of the key challenges in visual Reinforcement Learning (RL) is to learn policies that can generalize to unseen environments. Recently, data augmentation techniques aiming at enhancing data diversity have demonstrated proven performance in improving the generalization ability of learned policies. However, due to the sensitivity of RL training, naively applying data augmentation, which transforms each pixel in a task-agnostic manner, may suffer from instability and damage the sample efficiency, thus further exacerbating the generalization performance. At the heart of this phenomenon is the diverged action distribution and high-variance value estimation in the face of augmented images. To alleviate this issue, we propose Task-aware Lipschitz Data Augmentation (TLDA) for visual RL, which explicitly identifies the task-correlated pixels with large Lipschitz constants, and only augments the task-irrelevant pixels for stability. We verify the effectiveness of our approach on DeepMind Control suite, CARLA and DeepMind Manipulation tasks. The extensive empirical results show that TLDA improves both sample efficiency and generalization; it outperforms previous state-of-the-art methods across 3 different visual control benchmarks.

Safe reinforcement learning aims to learn the optimal policy while satisfying safety constraints, which is essential in real-world applications. However, current algorithms still struggle for efficient policy updates with hard constraint satisfaction. In this paper, we propose Penalized Proximal Policy Optimization (P3O), which solves the cumbersome constrained policy iteration via a single minimization of an equivalent unconstrained problem. Specifically, P3O utilizes a simple yet effective penalty approach to eliminate cost constraints and removes the trust-region constraint by the clipped surrogate objective. We theoretically prove the exactness of the penalized method with a finite penalty factor and provide a worst-case analysis for approximate error when evaluated on sample trajectories. Moreover, we extend P3O to more challenging multi-constraint and multi-agent scenarios which are less studied in previous work. Extensive experiments show that P3O outperforms state-of-the-art algorithms with respect to both reward improvement and constraint satisfaction on a set of constrained locomotive tasks.

The style-based generator architectures (e.g. StyleGAN v1, v2) largely promote the controllability and explainability of Generative Adversarial Networks (GANs). Many researchers have applied the pretrained style-based generators to image manipulation and video editing by exploring the correlation between linear interpolation in the latent space and semantic transformation in the synthesized image manifold. However, most previous studies focused on manipulating separate discrete attributes, which is insufficient to animate a still image to generate videos with complex and diverse poses and expressions. In this work, we devise a dual distillation strategy (D2Animator) for generating animated high-resolution face videos conditioned on identities and poses from different images. Specifically, we first introduce a Clustering-based Distiller (CluDistiller) to distill diverse interpolation directions in the latent space, and synthesize identity-consistent faces with various poses and expressions, such as blinking, frowning, looking up/down, etc. Then we propose an Augmentation-based Distiller (AugDistiller) that learns to encode arbitrary face deformation into a combination of interpolation directions via training on augmentation samples synthesized by CluDistiller. Through assembling the two distillation methods, D2Animator can generate high-resolution face animation videos without training on video sequences. Extensive experiments on self-driving, cross-identity and sequence-driving tasks demonstrate the superiority of the proposed D2Animator over existing StyleGAN manipulation and face animation methods in both generation quality and animation fidelity.

In unsupervised domain adaptation (UDA), a classifier for the target domain is trained with massive true-label data from the source domain and unlabeled data from the target domain. However, collecting true-label data in the source domain can be expensive and sometimes impractical. Compared to the true label (TL), a complementary label (CL) specifies a class that a pattern does not belong to, and hence, collecting CLs would be less laborious than collecting TLs. In this article, we propose a novel setting where the source domain is composed of complementary-label data, and a theoretical bound of this setting is provided. We consider two cases of this setting: one is that the source domain only contains complementary-label data [completely complementary UDA (CC-UDA)] and the other is that the source domain has plenty of complementary-label data and a small amount of true-label data [partly complementary UDA (PC-UDA)]. To this end, a complementary label adversarial network (CLARINET) is proposed to solve CC-UDA and PC-UDA problems. CLARINET maintains two deep networks simultaneously, with one focusing on classifying the complementary-label source data and the other taking care of the source-to-target distributional adaptation. Experiments show that CLARINET significantly outperforms a series of competent baselines on handwritten digit-recognition and object-recognition tasks.

The attitude control of an underactuated reusable launch vehicle (RLV) in the reentry phase involving nonminimum phase problem and control input constraints is investigated in this article. To address the nonminimum phase problem, an approach combining output redefinition and robust backstepping is proposed, where a synthetic output is constructed using the combination of the original output and the internal states to obtain stable zero dynamics, and then robust backstepping is performed on the new output. Besides, the ideal internal dynamics are obtained by using optimal bounded inversion, which are incorporated into the controller as the reference trajectories for the internal states to improve the output tracking accuracy. To cope with the control input constraints, a simple and useful anti-windup strategy is proposed by using feedback error clipping, which is shown to be very effective in mitigating control input saturation. Numerical simulations are given to validate the effectiveness of the proposed method.

Unsupervised domain adaptation (UDA) aims to train a target classifier with labeled samples from the source domain and unlabeled samples from the target domain. Classical UDA learning bounds show that target risk is upper bounded by three terms: source risk, distribution discrepancy, and combined risk. Based on the assumption that the combined risk is a small fixed value, methods based on this bound train a target classifier by only minimizing estimators of the source risk and the distribution discrepancy. However, the combined risk may increase when minimizing both estimators, which makes the target risk uncontrollable. Hence the target classifier cannot achieve ideal performance if we fail to control the combined risk. To control the combined risk, the key challenge takes root in the unavailability of the labeled samples in the target domain. To address this key challenge, we propose a method named E-MixNet. E-MixNet employs enhanced mixup, a generic vicinal distribution, on the labeled source samples and pseudo-labeled target samples to calculate a proxy of the combined risk. Experiments show that the proxy can effectively curb the increase of the combined risk when minimizing the source risk and distribution discrepancy. Furthermore, we show that if the proxy of the combined risk is added into loss functions of four representative UDA methods, their performance is also improved.

The CNN-based image inpainting methods have achieved promising performance because of its outstanding semantic understanding and reasoning potentialities. However, previous works could not get satisfied results in some situations because information is not fully explored. In this paper, we propose a new method by combining three innovative ideas. First, to increase the diversity of the semantic information obtained by the network in image synthesis, we propose a multiple hidden space perceptual (MHSP) loss, which extracts high-level features from multiple pre-trained autoencoders. Second, we adopt an adaptive iterative reasoning (AIR) stategy to reduce the calculations under small-hole circumstances while ensuring the performance in large-hole circumstances. Third, we find that color inconsistencies occasionally occurred in the final image merging process, so we add a novel interval maximum saturation (IMS) loss to the final loss function. Experiments on the benchmark datasets show our method performs favorably against state-of-the-art approaches. Code is made publicly available at: https://github.com/IC-LAB/adaptive_iterative_inpainting.

The next-item recommendation has been in the central of interest in real-world applications such as e-commerce. However, it is challenging to infer what a user may purchase next due the complex interactions in the historical sessions and the changing semantics of an item over time. Most existing methods employ separate models to generate the general preference and the sequential patterns for the next-item recommendation without considering the interactions between the two factors or use a simple linear combination of the two factors. In this paper, we propose a deep adaptable co-embedding neural network (ACENet) to address these limitations. ACENet not only adaptably balances the combination of general preference and sequential patterns but also introduces dynamic attention for each factor in hybrid representations. Extensive experiments on two real-world datasets show the superiority of ACENet compared with other state-of-the-art methods.

Portrait animation, which aims to animate a still portrait to life using poses extracted from target frames, is an important technique for many real-world entertainment applications. Although recent works have achieved highly realistic results on synthesizing or controlling human head images, the puppeteering of arbitrary portraits is still confronted by the following challenges: 1) identity/personality mismatch; 2) training data/domain limitations; and 3) low-efficiency in training/fine-tuning. In this paper, we devised a novel two-stage framework called PuppeteerGAN for solving these challenges. Specifically, we first learn identity-preserved semantic segmentation animation which executes pose retargeting between any portraits. As a general representation, the semantic segmentation results could be adapted to different datasets, environmental conditions or appearance domains. Furthermore, the synthesized semantic segmentation is filled with the appearance of the source portrait. To this end, an appearance transformation network is presented to produce fidelity output by jointly considering the wrapping of semantic features and conditional generation. After training, the two networks can directly perform end-to-end inference on unseen subjects without any retraining or fine-tuning. Extensive experiments on cross-identity/domain/resolution situations demonstrate the superiority of the proposed PuppetterGAN over existing portrait animation methods in both generation quality and inference speed.

Factorization Machines (FMs) refer to a class of general predictors working with real valued feature vectors, which are well-known for their ability to estimate model parameters under significant sparsity and have found successful applications in many areas such as the click-through rate (CTR) prediction. However, standard FMs only produce a single fixed representation for each feature across different input instances, which may limit the CTR model’s expressive and predictive power. Inspired by the success of Input-aware Factorization Machines (IFMs), which aim to learn more flexible and informative representations of a given feature according to different input instances, we propose a novel model named Dual Input-aware Factorization Machines (DIFMs) that can adaptively reweight the original feature representations at the bit-wise and vector-wise levels simultaneously. Furthermore, DIFMs strategically integrate various components including Multi-Head Self-Attention, Residual Networks and DNNs into a unified end-to-end model. Comprehensive experiments on two real-world CTR prediction datasets show that the DIFM model can outperform several state-of-the-art models consistently.

In unsupervised domain adaptation (UDA), classifiers for the target domain are trained with massive true-label data from the source domain and unlabeled data from the target domain. However, it may be difficult to collect fully-true-label data in a source domain given limited budget. To mitigate this problem, we consider a novel problem setting where the classifier for the target domain has to be trained with complementary-label data from the source domain and unlabeled data from the target domain named budget-friendly UDA (BFUDA). The key benefit is that it is much less costly to collect complementary-label source data (required by BFUDA) than collecting the true-label source data (required by ordinary UDA). To this end, complementary label adversarial network (CLARINET) is proposed to solve the BFUDA problem. CLARINET maintains two deep networks simultaneously, where one focuses on classifying complementary-label source data and the other takes care of the source-to-target distributional adaptation. Experiments show that CLARINET significantly outperforms a series of competent baselines.

Scene perceiving and understanding tasks including depth estimation, visual odometry (VO) and camera relocalization are fundamental for applications such as autonomous driving, robots and drones. Driven by the power of deep learning, significant progress has been achieved on individual tasks but the rich correlations among the three tasks are largely neglected. In previous studies, VO is generally accurate in local scope yet suffers from drift in long distances. By contrast, camera relocalization performs well in the global sense but lacks local precision. We argue that these two tasks should be strategically combined to leverage the complementary advantages, and be further improved by exploiting the 3D geometric information from depth data, which is also beneficial for depth estimation in turn. Therefore, we present a collaborative learning framework, consisting of DepthNet, LocalPoseNet and GlobalPoseNet with a joint optimization loss to estimate depth, VO and camera localization unitedly. Moreover, the Geometric Attention Guidance Model is introduced to exploit the geometric relevance among three branches during learning. Extensive experiments demonstrate that the joint learning scheme is useful for all tasks and our method outperforms current state-of-the-art techniques in depth estimation and camera relocalization with highly competitive performance in VO.

This paper proposes a multi-task control strategy for a quadruped robot named THU-QUAD II. The mechanical design of the robot ensures a wide range of motion for all joints, which allows it to stand and walk like a mammal as well as sprawl to the ground and crawl like a reptile. Five basic leg configurations are defined for the robot, including four mammal-type configurations with bidirectional knees and one sprawling-type configuration. A multi-task control framework is developed by combining configuration selection and gait planning. According to the locomotion environments, the robot can nimbly switch between different configurations, which gives it more flexibility when facing different tasks. For the mammal-type configuration, a parametric climbing gait is designed to traverse structural terrain. For the sprawling-type configuration, a crawling gait is designed to achieve robust locomotion on uneven terrain. Simulations and experiments show that the robot is capable to move on multiple challenging terrains, including doorsills, stairs, slopes, sand and stones. This paper demonstrates that even some challenging locomotion tasks can be achieved in a rather simple way without using complicated control algorithms, which suggests us to rethink about the leg configurations in designing quadruped robots.

In this paper, we propose an identity-preserving face hallucination (IPFH) method via deep reinforcement learning. Most existing methods ultra-resolve facial visual information in guidance of appearance similarity which rarely attend to recovering the semantic property, undermining further face analysis (e.g., recognition). We present a visual-semantic hallucinator relying on deep reinforcement learning to adaptively repair local details for the restoration of both identity and appearance characteristics. Specifically, we first capture the facial global topology structure to roughly recover the visual information with the pixel-wise similarity constraint. To super-resolve more photo-realistic faces, we explore the contextual interdependency to reconstruct facial local textural details (e.g., over-smoothed edges) with the constraints of visual and identity similarity. In terms of the visual similarity constraint, we develop the dual domain network with bidirectional consistency on both HR domain and LR domain to improve the appearance quality. Moreover, we introduce the identity constraint to encourage hallucinated faces to satisfy the identity property. Experimental results on several benchmarks demonstrate our method achieves promising performance on the recovery of visual and semantic information.

DBSCAN is a popular clustering algorithm that can discover clusters of arbitrary shapes with broad applications. However, DBSCAN is computationally expensive, as it performs range queries for all the points to determine their neighbors and grow the clusters. To address this problem, we propose a novel approximate density-based clustering algorithm named DBSVEC. DBSVEC introduces support vectors into density-based clustering, which allows performing range queries only on a small subset of points called the core support vectors. This technique significantly improves the efficiency while retaining high-quality cluster results. We evaluate the performance of DBSVEC via extensive experiments on real and synthetic datasets. The results show that DBSVEC is up to three orders of magnitude faster than DBSCAN. Compared with the state-of-the-art approximate density-based clustering methods, DBSVEC is up to two orders of magnitude faster, and the clustering results of DBSVEC are more similar to those of DBSCAN.

Factorization machines (FMs) are a class of general predictors working effectively with sparse data, which represents features using factorized parameters and weights. However, the accuracy of FMs can be adversely affected by the fixed representation trained for each feature, as the same feature is usually not equally predictive and useful in different instances. In fact, the inaccurate representation of features may even introduce noise and degrade the overall performance. In this work, we improve FMs by explicitly considering the impact of individual input upon the representation of features. We propose a novel model named \textit{Input-aware Factorization Machine} (IFM), which learns a unique input-aware factor for the same feature in different instances via a neural network. Comprehensive experiments on three real-world recommendation datasets are used to demonstrate the effectiveness and mechanism of IFM. Empirical results indicate that IFM is significantly better than the standard FM model and consistently outperforms four state-of-the-art deep learning based methods.

Next item recommendation is an important yet challenging task in real-world applications such as E-commerce. Since people often carry out a series of online shopping activities, in order to predict what a user may purchase next, it is essential to model the user's general taste as well as the sequential correlation between purchases. Existing models combine these two factors directly without considering the dynamic changes of a user's long-term and short-term preferences. Meanwhile, when a purchase session contains multiple items, not all of them have the same impact on the next item to purchase. To address these limitations, we propose a model that introduces hierarchical attention to dynamically balance between general taste (long-term preference) and sequential behavior (short-term preference). To weight individual items in the same session, we design a neural memory network with attention mechanism to learn the dynamic weights. Our model can adapt the embedding of each session as well as the embedding of long-term and short-term preferences. Extensive experiments on three real-world datasets show that our model significantly outperforms state-of-the-art methods based on commonly used evaluation metrics.

This paper proposes a robust fault detection and isolation (FDI) approach that combines active and passive robust FDI approaches. Standard active FDI approaches obtain robustness by using the unknown input observer (UIO) to decouple unknown inputs from residuals. Differently, standard passive FDI approaches achieve robustness by using the set theory to bound the effect of uncertain factors (disturbances and noises). In this paper, we combine the UIO-based and the set-based approaches to produce a mixed robust FDI, which can mitigate the disadvantages and exert the advantages of the two robust FDI approaches. In order to emphasize the role of set theory, the UIO design based on the set theory is named as the set-theoretic UIO (SUIO). A quadrotor subsystem is used to illustrate the effectiveness of the proposed FDI approach.

Network is a powerful paradigm for representing complex relationships and finding the community structure of networks can help people better understand the real world. Infomap, which employs the minimum description length as the optimization objective, is a competent algorithm for community structure analysis. In this paper, we propose a novel algorithm combining flow-based ensemble learning and Label Propagation Algorithm (LPA). Firstly, Infomap (without recursive steps) is incorporated into Core Groups Graph Clustering (CGGC), an ensemble learning framework for community detection. Next, the output of CGGC-Infomap is used as the input of LPA, which can make LPA converge to highly stable clustering results. Experimental studies show that our algorithm can achieve better performance in terms of Normalized Mutual Information (NMI) and requires less memory than the original Infomap algorithm. Our method also features good parallelism, making it potentially more suitable for processing large scale networks.

Clustering data streams has become a hot topic in the era of big data. Driven by the ever increasing volume, velocity and variety of data, more efficient algorithms for clustering large-scale complex data streams are needed. In this paper, we present a parallel algorithm called PaStream, which is based on advanced Graphics Processing Unit (GPU) and follows the online-offline framework of CluStream. Our approach can achieve hundreds of times speedup on high-speed and high-dimensional data streams compared with CluStream. It can also discover clusters with arbitrary shapes and handle outliers properly. The efficiency and scalability of PaStream are demonstrated through comprehensive experiments on synthetic and standard benchmark datasets with various problem factors.

ISODATA is a well-known clustering algorithm used in various areas. It employs a heuristic strategy allowing the clusters based on the nearest neighbor rule to split and merge as appropriate. However, since the volume of the data to be clustered in real world is growing continuously, the efficiency of serial ISODATA has become a serious practical issue. The GPU (Graphics Processing Unit) is an emerging high performance computing platform due to its highly parallel multithreaded architecture. In this paper, we propose an efficient parallel ISODATA algorithm based on the latest Kepler GPUs and the dynamic parallelism feature in CUDA (Compute Unified Device Architecture). Performance study shows that our parallel ISODATA can achieve promising speedup ratios and features favorable scalability compared to the original algorithm.

This paper presents a Kinect based hand gesture recognition system that can effectively recognize both one-hand and two-hand gestures. It is robust against the disturbance of complex background and objects such as the faces and hands of other people by exploiting the depth information and carefully choosing the region of interest (ROI) in the process of tracking. The recognition module is implemented using template matching and other light weight techniques to reduce the computational complexity. In the experiments, this system is tested on real world tasks from controlling the slide show in PowerPoint to playing the highly intense racing video game Need for Speed. The practical performance confirms that our system is both effective in terms of robustness and versatility and efficient for various real-time applications.

This paper presents a comprehensive review of Cyberinfrastructure (CI), an emerging collaborative research environment, including its representative applications in four science communities around the world. An in-depth analysis is also conducted to reveal the key functions and desired features that can be expected from modern CI systems.

This paper addresses the challenges in detecting the potential correlation between numerical data streams, which facilitates the research of data stream mining and pattern discovery. We focus on local correlation with delay, which may occur in burst at different time in different streams, and last for a limited period. The uncertainty on the correlation occurrence and the time delay make it difficult to monitor the correlation online. Furthermore, the conventional correlation measure lacks the ability of reflecting visual linearity, which is more desirable in reality. This paper proposes effective methods to continuously detect the correlation between data streams. Our approach is based on the Discrete Fourier Transform to make rapid cross-correlation calculation with time delay allowed. In addition, we introduce a shape-based similarity measure into the framework, which refines the results by representative trend patterns to enhance the significance of linearity. The similarity of proposed linear representations can quickly estimate the correlation, and the window sliding strategy in segment level improves the efficiency for online detection. The empirical study demonstrates the accuracy of our detection approach, as well as more than $30\%$ improvement of efficiency.

Emotion words play a vital role in many sentiment analysis tasks. Previous research uses sentiment dictionary to detect the subjectivity or polarity of words. In this paper, we dive into Emotion-Inducing Keywords (EIK), which refers to the words in use that convey emotion. We first analyze an emotion corpus to explore the pragmatic aspects of EIK. Then we design an effective framework for automatically detecting EIK in sentences by utilizing linguistic features and context information. Our system outperforms traditional dictionary-based methods dramatically in increasing Precision, Recall and F1-score.

Trajectory sharing and searching have received significant attentions in recent years. In this paper, we propose and investigate a novel problem called User Oriented Trajectory Search (UOTS) for trip recommendation. In contrast to conventional trajectory search by locations (spatial domain only), we consider both spatial and textual domains in the new UOTS query. Given a trajectory data set, the query input contains a set of intended places given by the traveler and a set of textual attributes describing the traveler's preference. If a trajectory is connecting/close to the specified query locations, and the textual attributes of the trajectory are similar to the traveler'e preference, it will be recommended to the traveler for reference. This type of queries can bring significant benefits to travelers in many popular applications such as trip planning and recommendation.

How to build a compact and informative training data set autonomously is crucial for many real-world learning tasks, especially those with large amount of unlabeled data and high cost of labeling. Active learning aims to address this problem by asking queries in a smart way. Two main scenarios of querying considered in the literature are query synthesis and pool-based sampling. Since in many cases synthesized queries are meaningless or difficult for human to label, more efforts have been devoted to pool-based sampling in recent years. However, in pool-based active learning, querying requires evaluating every unlabeled data point in the pool, which is usually very time-consuming. By contrast, query synthesis has clear advantage on querying time, which is independent of the pool size. In this paper, we propose a novel framework combining query synthesis and pool-based sampling to accelerate the learning process and overcome the current limitation of query synthesis. The basic idea is to select the data point nearest to the synthesized query as the query point. We also provide two simple strategies for synthesizing informative queries. Moreover, to further speed up querying, we employ clustering techniques on the whole data set to construct a representative unlabeled data pool based on cluster centers. Experiments on several real-world data sets show that our methods have distinct advantages in time complexity and similar performance compared to pool-based uncertainty sampling methods.

Existing attempts to improve the performance of AdaBoost on imbalanced datasets have largely been focused on modifying its weight updating rule or incorporating sampling or cost sensitive learning techniques. In this paper, we propose to tackle the challenge from a novel perspective. Initially, the dataset is over-sampled and the standard AdaBoost is applied to create a series of base classifiers. Next, the weights of the classifiers are further retrained by Genetic Algorithms (GAs) or comparable optimization techniques where more targeted performance measures such as G-mean and F-measure can be directly used as the objective function. Consequently, unlike other indirect solutions, this sampling + reweighting strategy can purposefully tune AdaBoost towards a certain performance measure of interest with only moderate computational overhead. Experimental results on ten benchmark datasets show that this strategy can reliably boost the performance of AdaBoost and has consistent superiority over EasyEnsemble, which is a competent ensemble method for class imbalance learning.

Traditional network processors (NPs) adopt pull model, where NP cores pull packet data from external memory to local memory, triggered by cache miss or fetch instructions. Due to the long latency of data fetching, hardware multithreading is typically used to reduce the waiting time. Multithreading incurs context switch overhead, leading to inefficiency in payload processing applications. We propose a push model for future NP's architectural design to increase throughput and decrease processing delay. A hardware push unit helps to move the segments of a packet to a core's local memory to reduce hardware thread switching. Theoretical analyses are given to compare the pull and push model's performance. Further, we selected our FPGA based THNPU NP platform for verification. Experimental results indicate that the push model not only improves the system throughput, but also reduces the delay, with only a fraction of logic gate increase.

With the fast development of Internet, the size of routing tables in the backbone routers keeps a rapid growth in recent years. An effective solution to control the memory occupation of the ever-increased huge routing table is the Forwarding Information Base (FIB) compression. Existing optimal FIB compression algorithm ORTC suffers from high computational complexity and poor update performance, due to the loss of essential structure information during its compression process. To address this problem, we present two suboptimal FIB compression algorithms — EAR-fast and EAR-slow, respectively, based on our proposed Election and Representative (EAR) algorithm which is an optimal FIB compression algorithm. The two suboptimal algorithms preserve the structure information, and support fast incremental updates while reducing computational complexity. Experiments on an 18-month real data set show that compared with ORTC, the proposed EAR-fast algorithm requires only 9.8% compression time and 37.7% memory space, but supports faster update while prolonging the recompression interval remarkably. All these performance advantages come at a cost of merely a 1.5% loss in compression ratio compared with the theoretical optimal ratio.

Nonnegative matrix factorization (NMF) has become a popular dimension-reduction method and has been widely applied to image processing and pattern recognition problems. However, conventional NMF learning methods require the entire dataset to reside in the memory and thus cannot be applied to large-scale or streaming datasets. In this paper, we propose an efficient online RSA-NMF algorithm (OR-NMF) that learns NMF in an incremental fashion and thus solves this problem. In particular, OR-NMF receives one sample or a chunk of samples per step and updates the bases via robust stochastic approximation. Benefitting from the smartly chosen learning rate and averaging technique, OR-NMF converges at the rate of in each update of the bases. Furthermore, we prove that OR-NMF almost surely converges to a local optimal solution by using the quasi-martingale. By using a buffering strategy, we keep both the time and space complexities of one step of the OR-NMF constant and make OR-NMF suitable for large-scale or streaming datasets. Preliminary experimental results on real-world datasets show that OR-NMF outperforms the existing online NMF (ONMF) algorithms in terms of efficiency. Experimental results of face recognition and image annotation on public datasets confirm the effectiveness of OR-NMF compared with the existing ONMF algorithms.

In this paper, we propose and investigate a novel spatial query called Reverse Path Nearest Neighbor (R-PNN) search to find the most accessible locations in road networks. Given a trajectory data-set and a list of location candidates specified by users, if a location o is the Path Nearest Neighbor (PNN) of k trajectories, the influence-factor of o is defined as k and the R-PNN query returns the location with the highest influence-factor. The R-PNN query is an extension of the conventional Reverse Nearest Neighbor (RNN) search. It can be found in many important applications such as urban planning, facility allocation, traffic monitoring, etc. To answer the R-PNN query efficiently, an effective trajectory data pre-processing technique is conducted in the first place. We cluster the trajectories into several groups according to their distribution. Based on the grouped trajectory data, a two-phase solution is applied. First, we specify a tight search range over the trajectory and location data-sets. The efficiency study reveals that our approach defines the minimum search area. Second, a series of optimization techniques are adopted to search the exact PNN for trajectories in the candidate set. By combining the PNN query results, we can retrieve the most accessible locations. The complexity analysis shows that our solution is optimal in terms of time cost. The performance of the proposed R-PNN query processing is verified by extensive experiments based on real and synthetic trajectory data in road networks.

Nonnegative matrix factorization (NMF) has become a popular data-representation method and has been widely used in image processing and pattern-recognition problems. This is because the learned bases can be interpreted as a natural parts-based representation of data and this interpretation is consistent with the psychological intuition of combining parts to form a whole. For practical classification tasks, however, NMF ignores both the local geometry of data and the discriminative information of different classes. In addition, existing research results show that the learned basis is unnecessarily parts-based because there is neither explicit nor implicit constraint to ensure the representation parts-based. In this paper, we introduce the manifold regularization and the margin maximization to NMF and obtain the manifold regularized discriminative NMF (MD-NMF) to overcome the aforementioned problems. The multiplicative update rule (MUR) can be applied to optimizing MD-NMF, but it converges slowly. In this paper, we propose a fast gradient descent (FGD) to optimize MD-NMF. FGD contains a Newton method that searches the optimal step length, and thus, FGD converges much faster than MUR. In addition, FGD includes MUR as a special case and can be applied to optimizing NMF and its variants. For a problem with 165 samples in R(1600), FGD converges in 28 s, while MUR requires 282 s. We also apply FGD in a variant of MD-NMF and experimental results confirm its efficiency. Experimental results on several face image datasets suggest the effectiveness of MD-NMF.

In this paper, we present a non-negative patch alignment framework (NPAF) to unify popular non-negative matrix factorization (NMF) related dimension reduction algorithms. It offers a new viewpoint to better understand the common property of different NMF algorithms. Although multiplicative update rule (MUR) can solve NPAF and is easy to implement, it converges slowly. Thus, we propose a fast gradient descent (FGD) to overcome the aforementioned problem. FGD uses the Newton method to search the optimal step size, and thus converges faster than MUR. Experiments on synthetic and real-world datasets confirm the efficiency of FGD compared with MUR for optimizing NPAF. Based on NPAF, we develop non-negative discriminative locality alignment (NDLA). Experiments on face image and handwritten datasets suggest the effectiveness of NDLA in classification tasks and its robustness to image occlusions, compared with representative NMF-related dimension reduction algorithms.

More complicated computational tasks are posed to the network equipments, such as Deep packet inspection (DPI) for network security check and network coding to achieve efficient multicast, etc. These complicated applications need processors to process the whole packet payload, potentially causing low throughput and long latency due to the large access delay to external memories. The behind hint lies that we can get the packet-processor/thread pair binding information in advance from the front-end dispatching component before the packet will be actually processed by cores. This interesting observation enables us design a new architecture of memory access for packet processors instead of the traditional model. In this paper we explore to apply push model to packet processors. The push model makes the data being pushed into the local memory/on-chip L1 cache in an on-demand and fine granularity manner ahead of being asked by running instructions, making a core always feels getting its data from the local memory/L1 cache instead of fetching them from the external memory in pull model. In order to verify the effectiveness, we design and implement the push model with the Intel IXP2850, and then conduct experiments to show the performance of push model in the IXP2850 simulator compared with the pull model. Simulation results indicate that applying push model to packet processors could improve the system throughput and reduce the packet processing latency and reducing required number of hardware threads.

This paper presents some new analytical results on the continuous Univariate Marginal Distribution Algorithm (UMDAC), which is a well known Estimation of Distribution Algorithm based on Gaussian distributions. As the extension of the current theoretical work built on the assumption of infinite populations, the convergence behavior of UMDAC with finite populations is formally analyzed. We show both analytically and experimentally that, on flat landscapes, the Gaussian model in UMDAC tends to collapse with high probability, which is an important fact that is not well understood before.

Success rate is a commonly adopted performance criterion for evaluating Evolutionary Algorithms due to their inherent randomness. However, the classical large-sample binomial test based on normal distributions is only valid with a relatively large number of trials, which may not be feasible when experimental studies are very time consuming or expensive. In this paper, we give an alternative statistical test, which is suitable for situations where results from only a small number of trials are available.

Given a set of sparsely distributed sensors in the Euclidean plane, a mobile robot is required to visit all sensors to download the data and finally return to its base. The effective range of each sensor is specified by a disk, and the robot must at least reach the boundary to start communication. The primary goal of optimization in this scenario is to minimize the traveling distance by the robot. This problem can be regarded as a special case of the traveling salesman problem with neighborhoods (TSPN), which is known to be NP-hard. In this paper, we present a novel TSPN algorithm for this class of TSPN, which can yield significantly improved results compared to the latest approximation algorithm.

This paper presents some initial attempts to mathematically model the dynamics of a continuous Estimation of Distribution Algorithm (EDA) based on a Gaussian distribution and truncation selection. Case studies are conducted on both unimodal and multimodal problems to highlight the effectiveness of the proposed technique and explore some important properties of the EDA. With some general assumptions, we show that, for 1D unimodal problems and with the (µ, λ) scheme: (1). The behaviour of the EDA is dependent only on the general shape of the test function, rather than its specific form; (2). When initialized far from the global optimum, the EDA has a tendency to converge prematurely; (3). Given a certain selection pressure, there is a unique value for the proposed amplification parameter that could help the EDA achieve desirable performance; for 1D multimodal problems: (1). The EDA could get stuck with the (µ, λ) scheme; (2). The EDA will never get stuck with the (µ +λ) scheme.

The research literature on metaheuristic and evolutionary computation has proposed a large number of algorithms for the solution of challenging real-world optimization problems. It is often not possible to study theoretically the performance of these algorithms unless significant assumptions are made on either the algorithm itself or the problems to which it is applied, or both. As a consequence, metaheuristics are typically evaluated empirically using a set of test problems. Unfortunately, relatively little attention has been given to the development of methodologies and tools for the large-scale empirical evaluation and/or comparison of metaheuristics. In this paper, we propose a landscape (test-problem) generator that can be used to generate optimization problem instances for continuous, bound-constrained optimization problems. The landscape generator is parameterized by a small number of parameters, and the values of these parameters have a direct and intuitive interpretation in terms of the geometric features of the landscapes that they produce. An experimental space is defined over algorithms and problems, via a tuple of parameters for any specified algorithm and problem class (here determined by the landscape generator). An experiment is then clearly specified as a point in this space, in a way that is analogous to other areas of experimental algorithmics, and more generally in experimental design. Experimental results are presented, demonstrating the use of the landscape generator. In particular, we analyze some simple, continuous estimation of distribution algorithms, and gain new insights into the behavior of these algorithms using the landscape generator.

Choosing the best parameter setting is a well-known important and challenging task in evolutionary algorithms (EAs). As one of the earliest parameter tuning techniques, the meta-EA approach regards each parameter as a variable and the performance of algorithm as the fitness value and conducts searching on this landscape using various genetic operators. However, there are some inherent issues in this method. For example, some algorithm parameters are generally not searchable because it is difficult to define any sensible distance metric on them. In this paper, a novel approach is proposed by combining the meta-EA approach with a method called racing, which is based on the statistical analysis of algorithm performance with different parameter settings. A series of experiments are conducted to show the reliability and efficiency of this hybrid approach in tuning genetic algorithms (GAs) on two benchmark problems.

A comprehensive set of experiments was conducted with a continuous EDA on 25 test problems provided in the real-parameter optimization special session. It is expected that the results presented here could be used to gain some deeper understanding of the performance of the EDA as well as facilitate the comparison across different algorithms.

The development of Estimation of Distribution Algorithms (EDAs) has largely been driven by using more and more complex statistical models to approximate the structure of search space. However, there are still problems that are difficult for EDAs even with models capable of capturing high order dependences. In this paper, we show that diversity maintenance plays an important role in the performance of EDAs. A continuous EDA based on the Cholesky decomposition is tested on some well-known difficult benchmark problems to demonstrate how different diversity maintenance approaches could be applied to substantially improve its performance.

This paper introduces the design of superconductive magnet configurations in Magnetic Resonance Imaging (MRI) systems as a challenging real-world problem for Evolutionary Algorithms (EAs). Analysis of the problem structure is conducted using a general statistical method, which could be easily applied to other problems. The results suggest that the problem is highly multimodal and likely to present a significant challenge for many algorithms. Through a series of preliminary experiments, a continuous Estimation of Distribution Algorithm (EDA) is shown to be able to generate promising designs with a small computational effort. The importance of utilizing problem-specific knowledge and the ability of an algorithm to capture dependencies in solving complex real-world problems is also highlighted.

As an alternative to traditional evolutionary algorithms (EAs), population-based incremental learning (PBIL) maintains a probabilistic model of the best individual(s). Originally, PBIL was applied in binary search spaces. Recently, some work has been done to extend it to continuous spaces. In this paper, we review two such extensions of PBIL. An improved version of the PBIL based on Gaussian model is proposed that combines two main features: a new updating rule that takes into account all the individuals and their fitness values and a self-adaptive learning rate parameter. Furthermore, a new continuous PBIL employing a histogram probabilistic model is proposed. Some experiments results are presented that highlight the features of the new algorithms.

In this paper, we address some issue related to evaluating and testing evolutionary algorithms. A landscape generator based on Gaussian functions is proposed for generating a variety of continuous landscapes as fitness functions. Through some initial experiments, we illustrate the usefulness of this landscape generator in testing evolutionary algorithms.