Real-world time series data are usually multivariate with complex channel relations. Some channels are highly related, while others are with limited correlation. Channel independence has shown great significance in multivariate time series forecasting to capture better individual channel characteristics. In order to learn channel patterns better, we propose a channel-independent Refiner as a plug-and-play module. Specifically, we devise a channel-wise Refiner connected to the output of existing methods. Leveraging the concatenation of original input and the coarse prediction from the base model, the Refiner produces a better estimation. Our Refiner also benefits from the channel-independent design and a post-training strategy, achieving significant improvement over base models. Extensive experiments on iTransformer, FEDformer, Autoformer, FreTS, DLinear and TSMixer demonstrate that our Refiner reduces forecasting errors on both Transformer-based and MLP-based models in over 90% of the experimental settings. Our Refiner combined with iTransformer establishes the new state-of-the-art.

ABSTRACT Images captured by underwater robots often suffer from issues such as blurring and colour distortion, which hinder effective feature extraction and target recognition in underwater environments. To address these challenges, this paper proposes a novel underwater image enhancement method based on generative adversarial networks (GANs), termed multiple colour space underwater generative adversarial network (MCS‐UGAN). The proposed method is built upon a GAN framework, consisting of a generator and a discriminator. The generator comprises two main modules: a deblurring module and a colour correction module. The deblurring module innovatively incorporates an efficient multi‐scale feature extraction technique and an attention mechanism, which enhances object contours while preserving fine image details. The colour correction module integrates residual blocks into the U‐Net architecture, effectively mitigating the problems of gradient vanishing and explosion during backpropagation in underwater image enhancement networks, thereby enhancing the network's feature learning capability. This design corrects colour distortions while preserving edge information in the image. The discriminator adopts the PatchGAN structure, which focuses on the local regions of the image, significantly improving the generator's ability to restore high‐frequency details and thus enhancing the quality of the generated images. Experimental results on benchmark datasets demonstrate that, compared to existing methods, MCS‐UGAN achieves superior performance in terms of peak signal‐to‐noise ratio, structural similarity index measure, underwater image quality measure, and underwater colour image quality evaluation, with average values of 26.24, 0.91, 3.13, and 0.64, respectively. Results from real‐world applications further show that MCS‐UGAN effectively increases the number of extracted corner points, validating its practicality and effectiveness. The code is available at https://github.com/invincibility6/MCS‐UGAN.git

Continual learning (CL) aims to learn a sequence of tasks without forgetting. Numerous efforts have been made to tackle CL including data-centric, model-centric, and algorithm-centric methods. The more information the algorithm can obtain from previous tasks, e.g., the training data, the easier the CL task will be. However, few studies focus on the most difficult setting, i.e., dark CL (DCL) where only the model of the last task can be obtained. DCL is a typical setting in real-world applications, e.g., Cl tasks based on the models trained on private or privileged data. For solving DCL, we propose a novel recursive generalization bound, which can also be applied to arbitrary Traditional CL (TCL). To minimize the bound proposed, we propose a novel method, i.e., conditional past experience generation (CPEG), which reconstructs the previous conditional training data in the DCL setting. In the experiment, we apply CPEG to a wide range of benchmarks. The experimental results show that CPEG significantly reduces forgetting. On the other hand, CPEG can be used as a regularization term for any CL baseline. We also conduct experiments on the TCL setting. The performance of almost all baselines is improved, especially for the most difficult class-incremental tasks.

Unsupervised domain adaptation (UDA) has attracted much attention among those seeking to transfer a model from a labeled source domain to an unlabeled target domain. Many effective algorithms for single target domain adaptation (STDA) have been designed, however, STDA cannot satisfy the scenarios of transferring simultaneously to multiple target domains or transferring to a blending target domain. This paper proposes a novel discriminative mutual learning method for multi-target domain adaptation covering both blending target domain adaptation (BTDA) and multiple target domain adaptation (MTDA). Two key points are considered in the proposed method: one is to learn discriminative features for better prediction, and the other is to self-train the model with pseudo-labeled target data based on distance information. These two aspects are integrated through a mutual learning strategy via two different classifiers. According to extensive experiments on three domain adaptation benchmarks, the proposed method demonstrates the state-of-the-art performance in both BTDA and MTDA settings.

Existing unsupervised domain adaptation (UDA) studies focus on transferring knowledge in an offline manner. However, many tasks involve online requirements, especially in real-time systems. In this paper, we discuss Online UDA (OUDA) which assumes that the target samples are arriving sequentially as a small batch. OUDA tasks are challenging for prior UDA methods since online training suffers from catastrophic forgetting which leads to poor generalization. Intuitively, a good memory is a crucial factor in the success of OUDA. We formalize this intuition theoretically with a generalization bound where the OUDA target error can be bounded by the source error, the domain discrepancy distance, and a novel metric on forgetting in continuous online learning. Our theory illustrates the tradeoffs inherent in learning and remembering representations for OUDA. To minimize the proposed forgetting metric, we propose a novel source feature distillation (SFD) method which utilizes the source-only model as a teacher to guide the online training. In the experiment, we modify three UDA algorithms, i.e., DANN, CDAN, and MCC, and evaluate their performance on OUDA tasks with real-world datasets. By applying SFD, the performance of all baselines is significantly improved.

Recent studies explore how to obtain adversarial robustness for unsupervised domain adaptation (UDA). These efforts are however dedicated to achieving an optimal trade-off between accuracy and robustness on a given or seen target domain but ignore the robust generalization issue over unseen adversarial data. Consequently, degraded performance will be often observed when existing robust UDAs are applied to future adversarial data. In this work, we make a first attempt to address the robust generalization issue of UDA. We conjecture that the poor robust generalization of present robust UDAs may be caused by the large distribution gap among adversarial examples. We then provide an empirical and theoretical analysis showing that this large distribution gap is mainly owing to the discrepancy between feature-shift distributions. To reduce such discrepancy, a novel Anchored Feature-Shift Regularization (AFSR) method is designed with a certificated robust generalization bound. We conduct a series of experiments on benchmark UDA datasets. Experimental results validate the effectiveness of our proposed AFSR over many existing robust UDA methods.

In this work, we focus on supervised domain adaptation for object detection in few-shot loose annotation setting, where the source images are sufficient and fully labeled but the target images are few-shot and loosely annotated. As annotated objects exist in the target domain, instance level alignment can be utilized to improve the performance. Traditional methods conduct the instance level alignment by semantically aligning the distributions of paired object features with domain adversarial training. Although it is demonstrated that point-wise surrogates of distribution alignment provide a more effective solution in few-shot classification tasks across domains, this point-wise alignment approach has not yet been extended to object detection. In this work, we propose a method that extends the point-wise alignment from classification to object detection. Moreover, in the few-shot loose annotation setting, the background ROIs of target domain suffer from severe label noise problem, which may make the point-wise alignment fail. To this end, we exploit moving average centroids to mitigate the label noise problem of background ROIs. Meanwhile, we exploit point-wise alignment over instances and centroids to tackle the problem of scarcity of labeled target instances. Hence this method is not only robust against label noises of background ROIs but also robust against the scarcity of labeled target objects. Experimental results show that the proposed instance level alignment method brings significant improvement compared with the baseline and is superior to state-of-the-art methods.

Unsupervised domain adaptation (UDA) methods aim to transfer knowledge from a labeled source domain to an unlabeled target domain. Most existing UDA methods try to learn domain-invariant features so that the classifier trained by the source labels can automatically be adapted to the target domain. However, recent works have shown the limitations of these methods when label distributions differ between the source and target domains. Especially, in partial domain adaptation (PDA) where the source domain holds plenty of individual labels (private labels) not appeared in the target domain, the domain-invariant features can cause catastrophic performance degradation. In this paper, based on the originally favorable underlying structures of the two domains, we learn two kinds of target features, i.e., the source-approximate features and target-approximate features instead of the domain-invariant features. The source-approximate features utilize the consistency of the two domains to estimate the distribution of the source private labels. The target-approximate features enhance the feature discrimination in the target domain while detecting the hard (outlier) target samples. A novel Coupled Approximation Neural Network (CANN) has been proposed to co-train the source-approximate and target-approximate features by two parallel sub-networks without sharing the parameters. We apply CANN to three prevalent transfer learning benchmark datasets, Office-Home, Office-31, and Visda2017 with both UDA and PDA settings. The results show that CANN outperforms all baselines by a large margin in PDA and also performs best in UDA.

The latest heuristic for handling the domain shift in un-supervised domain adaptation tasks is to reduce the data distribution discrepancy using adversarial learning. Recent studies improve the conventional adversarial domain adaptation methods with discriminative information by integrating the classifier’s outputs into distribution divergence measurement. However, they still suffer from the equilibrium problem of adversarial learning in which even if the discriminator is fully confused, sufficient similarity between two distributions cannot be guaranteed. To overcome this problem, we propose a novel approach named feature gradient distribution alignment (FGDA)1. We demonstrate the rationale of our method both theoretically and empirically. In particular, we show that the distribution discrepancy can be reduced by constraining feature gradients of two domains to have similar distributions. Meanwhile, our method enjoys a theoretical guarantee that a tighter error upper bound for target samples can be obtained than that of conventional adversarial domain adaptation methods. By integrating the proposed method with existing adversarial domain adaptation models, we achieve state-of-the-art performance on two real-world benchmark datasets.

Unsupervised domain adaptation (UDA) methods are generally proposed based on the assumption that the source domain and the target domain share an identical group of classes. However, in transfer learning tasks in reality, the target domain often has fewer data with missing classes. Partial domain adaptation (PDA) allows the source domain to have un-shared categories. Anchor points are used to describe the easily identified target samples. It is observed that the shared classes tend to have more anchor points compared with the unshared classes and we introduce a novel progressive optimization method named Ebb PDA tasks. Ebb could resist the negative transfer caused by the category gap and can be applied to any domain adaptation model. Ebb picks the anchor points by analyzing the features of the base model, and it uses the class-wise distribution of anchor points to estimate the category gap. Then Ebb minimizes the errors of shared classes and corrects the error samples caused by blind alignment. To verify the effectiveness of the method, we apply Ebb to three PDA image classification tasks based on three widely used data sets, i.e, Office-Home, Office-31 and ImageCLEF-DA while using three state-of-the-art methods as the base models. The results show that Ebb brings a significant improvement in all tasks and the models optimized by Ebb have stable performance under a wide range of category gaps.

Unsupervised domain adaptation (UDA) is a challenging task characterized by unlabeled target data with domain discrepancy to labeled source data. Many methods have been proposed to learn domain invariant features by marginal distribution alignment, but they ignore the intrinsic structure within target domain, which may lead to insufficient or false alignment. Class-level alignment has been demonstrated to align the features of the same class between source and target domains. These methods rely extensively on the accuracy of predicted pseudo-labels for target data. Here, we develop a novel self-training method that focuses more on accurate pseudo-labels via a dual-consistency strategy involving modelling the intrinsic structure of the target domain. The proposed dual-consistency strategy first improves the accuracy of pseudo-labels through voting consistency, and then reduces the negative effects of incorrect predictions through structure consistency with the relationship of intrinsic structures across domains. Our method has achieved comparable performance to the state-of-the-arts on three standard UDA benchmarks.

Domain shift is regarded as a key factor affecting the robust-ness of many models. Recently, unsupervised auxiliary learning (e.g., input reconstruction) has been proposed to improve the model’s domain transferability and alleviate cross-domain performance degradation; however, in the paradigm of existing approaches, the features extracted from various tasks are shared, which mixes the domain-invariant features from the main task and domain-specific feature from the auxiliary task, leading to an imperfect learning. To solve this problem, we propose a novel unsupervised domain adaptation method - the Disentangled Reconstruction Neural Network (DRNN) - for cross-domain retina vessel segmentation. DRNN leverages two tandem nets and disentangles the domain-invariant features and the domain-specific features in the multi-task learning process. We perform extensive experiments on public retina datasets and our proposed DRNN outperforms the competitors by a significant margin to achieve state-of-the-art results pertaining to retina vessel segmentation.

It is important to estimate the performance gap of a given deep learning model on the target data set, since discrepancy or bias between source and target domains is a common and fundamental problem in the practice of machine learning techniques. Without any assumptions on data bias, such as label shift or covariate shift and without target data labels, we propose a robustness estimation method based on prediction consistency evaluation between source and target data in the neighborhood of the source samples. Considering outliers and whether the user provided model is fully trained, a variety of variant methods are also tried, including setting neighborhood threshold to average intra-class distance for each category and relative robustness. Furthermore, the time complexity of this method is O(nlogn), which is applicable for large datasets. Experiments on the handwritten digit recognition and Japanese handwriting recognition show that the proposed methods are effective.

This paper discusses the problem of local path planning in a static-obstacle environment by designing a PSO-based receding horizon control approach. In order to avoid obstacles, a virtual robot is first designed and moves along the boundary of obstacles. Then, in the framework of receding horizon control, a cost function is proposed where the virtual robot and the target position are integrated, which implies that mobile robots are controlled to keep a security distance and velocity consensus with virtual robots, and to move toward the target position. Next, the proposed cost function with constraints is processed by a particle swarm optimization (PSO) algorithm such that the PSO-based receding horizon control approach is developed. By solving the proposed cost function, a control sequence is obtained and then the first control input is used to enable the robot toward the target and avoid obstacles. Finally, the performance capabilities of the PSO-based receding horizon control approach are illustrated by simulation results.

This paper proposes an efficient wayfinding strategy for robot navigation in regionalized environments by designing a regionalized spatial knowledge model (RSK model) and a region-based wayfinding algorithm, i.e., a fine-to-coarse A* (FTC-A*) search algorithm. First, the RSK model, which imitates the representation of environments in the human brain, is presented to describe the search environments. The environments that are divided into regions are represented by a hierarchical nested structure where small regions are grouped together to form superordinate regions. Second, on the basis of the RSK model, an FTC-A* search algorithm is developed to plan the fine-to-coarse route. By making a fine planning to robot surroundings in vicinity, but a coarse planning to that at the distance, the FTC-A* algorithm can effectively reduce computational complexity, so as to enhance the efficiency of route search, and meanwhile makes robots to react quickly to user's commands, especially in large-scale environments. Finally, four exhaustive simulations and a physical experiment have been carried out to illustrate the feasibility and effectiveness of the proposed wayfinding strategy.

This paper presents a novel auction-based structure to multi-robot coalition formation problem. The structure, which is called multi-robot Coalition Structure Generation based on Credit Mechanism (CoSGCrM), contains a sub-optimal coalition member selection algorithm with an analysis of its soundness and completeness. A credit mechanism is introduced to reduce the complexity for the coalition leader in making a decision as well as to restrict the profit-oriented robot member in bidding for coalitions. Simulations are given to compare with first-price auction algorithm and the results show the viability of the proposed structure in both simple and complex tasks environments.

A simple and effective novel approach for scene understanding is addressed in this paper. Based on bag of visual words (BOVW) model, explicit semantics associated with the object image was embedded into visual words, and then various types of visual words integrated, and finally the SEVI-BOVW (semantics embedded and vocabulary integrated bag of visual words) model constructed. Mean Shift algorithm was employed to recognize local region image in scene. Compared with image understanding approaches presented in the literature, the proposed approach here can remove a classification or generative model during model training or testing. Objects category recognition can be determined by the number of class-specific semantic visual words, without complex reasoning. The effectiveness of the proposed approach has been demonstrated by the experimental results of scene understanding in a campus.

Web services technology is mainly designed for the interoperability among machines, whose inputs and outputs are defined in the form of XML schema without presentation information, hence it is unable to present the execution results of Web services to human beings directly and friendly. To address this problem, an ontology-based method and a rendering engine are proposed in this paper. In the method, domain ontology is utilized to annotate the types of inputs and outputs of a Web service. Templates are generated and stored according to each type described in the ontology. When rendering, the rendering engine first finds out the type of the execution results, and then retrieves corresponding templates with the URL of the type, finally recursively renders the execution results with the templates. In addition, a management interface is proposed for managing the templates.