Domain Adaptive Ensemble Learning

Domain generalization for image classification based on simplified self ensemble learning

Domain generalization seeks to acquire knowledge from limited source data and apply it to an unknown target domain. Current approaches primarily tackle this challenge by attempting to eliminate the differences between domains. However, as cross-domain data evolves, the discrepancies between domains grow increasingly intricate and difficult to manage, rendering effective knowledge transfer across multiple domains a persistent challenge. While existing methods concentrate on minimizing domain discrepancies, they frequently encounter difficulties in maintaining effectiveness when confronted with high data complexity. In this paper, we present an approach that transcends merely eliminating domain discrepancies by enhancing the model's adaptability to improve its performance in unseen domains. Specifically, we frame the problem as an optimization process with the objective of minimizing a weighted loss function that balances cross-domain discrepancies and sample complexity. Our proposed self-ensemble learning framework, which utilizes a single feature extractor, simplifies this process by alternately training multiple classifiers with shared feature extractors. The introduction of focal loss and complex sample loss weight further fine-tunes the model's sensitivity to hard-to-learn instances, enhancing generalization to difficult samples. Finally, a dynamic loss adaptive weighted voting strategy ensures more accurate predictions across diverse domains. Experimental results on three public benchmark datasets (OfficeHome, PACS, and VLCS) demonstrate that our proposed algorithm achieves an improvement of up to 3 . 38% over existing methods in terms of generalization performance, particularly in complex and diverse real-world scenarios, such as autonomous driving and medical image analysis. These results highlight the practical utility of our approach in environments where cross-domain generalization is crucial for system reliability and safety.

EEG-DG: A Multi-Source Domain Generalization Framework for Motor Imagery EEG Classification

Motorimagery EEG classification plays a crucial role in non-invasive Brain-Computer Interface (BCI) research. However, the performance of classification is affected by the non-stationarity and individual variations of EEG signals. Simply pooling EEG data with different statistical distributions to train a classification model can severely degrade the generalization performance. To address this issue, the existing methods primarily focus on domain adaptation, which requires access to the test data during training. This is unrealistic and impractical in many EEG application scenarios. In this paper, we propose a novel multi-source domain generalization framework called EEG-DG, which leverages multiple source domains with different statistical distributions to build generalizable models on unseen target EEG data. We optimize both the marginal and conditional distributions to ensure the stability of the joint distribution across source domains and extend it to a multi-source domain generalization framework to achieve domain-invariant feature representation, thereby alleviating calibration efforts. Systematic experiments conducted on a simulative dataset, BCI competition IV 2a, 2b, and OpenBMI datasets, demonstrate the superiority and competitive performance of our proposed framework over other state-of-the-art methods. Specifically, EEG-DG achieves average classification accuracies of 81.79% and 87.12% on datasets IV-2a and IV-2b, respectively, and 78.37% and 76.94% for inter-session and inter-subject evaluations on dataset OpenBMI, which even outperforms some domain adaptation methods.

Domain-guided conditional diffusion model for unsupervised domain adaptation

Limited transferability hinders the performance of a well-trained deep learning model when applied to new application scenarios. Recently, Unsupervised Domain Adaptation (UDA) has achieved significant progress in addressing this issue via learning domain-invariant features. However, the performance of existing UDA methods is constrained by the possibly large domain shift and limited target domain data. To alleviate these issues, we propose a Domain-guided Conditional Diffusion Model (DCDM), which generates high-fidelity target domain samples, making the transfer from source domain to target domain easier. DCDM introduces class information to control labels of the generated samples, and a domain classifier to guide the generated samples towards the target domain. Extensive experiments on various benchmarks demonstrate that DCDM brings a large performance improvement to UDA.

When Adversarial Training Meets Prompt Tuning: Adversarial Dual Prompt Tuning for Unsupervised Domain Adaptation

Unsupervised domain adaptation (UDA) aims to adapt models learned from a well-annotated source domain to a target domain, where only unlabeled samples are available. To this end, adversarial training is widely used in conventional UDA methods to reduce the discrepancy between source and target domains. Recently, prompt tuning has emerged as an efficient way to adapt large pre-trained vision-language models like CLIP to a variety of downstream tasks. In this paper, we present a novel method named Adversarial DuAl Prompt Tuning (ADAPT) for UDA, which employs text prompts and visual prompts to guide CLIP simultaneously. Rather than simply performing a joint optimization of text prompts and visual prompts, we integrate text prompt tuning and visual prompt tuning into a collaborative framework where they engage in an adversarial game: text prompt tuning focuses on distinguishing between source and target images, whereas visual prompt tuning seeks to align source and target domains. Unlike most existing adversarial training-based UDA approaches, ADAPT does not require explicit domain discriminators for domain alignment. Instead, the objective is effectively achieved at both global and category levels through modeling the joint probability distribution of images on domains and categories. Extensive experiments on four benchmark datasets demonstrate the effectiveness of our ADAPT method for UDA. We have released our code at https://github.com/Liuziyi1999/ADAPT.

Constrained Maximum Cross-Domain Likelihood for Domain Generalization

As a recent noticeable topic, domain generalization aims to learn a generalizable model on multiple source domains, which is expected to perform well on unseen test domains. Great efforts have been made to learn domain-invariant features by aligning distributions across domains. However, existing works are often designed based on some relaxed conditions which are generally hard to satisfy and fail to realize the desired joint distribution alignment. In this article, we propose a novel domain generalization method, which originates from an intuitive idea that a domain-invariant classifier can be learned by minimizing the Kullback-Leibler (KL)-divergence between posterior distributions from different domains. To enhance the generalizability of the learned classifier, we formalize the optimization objective as an expectation computed on the ground-truth marginal distribution. Nevertheless, it also presents two obvious deficiencies, one of which is the side-effect of entropy increase in KL-divergence and the other is the unavailability of ground-truth marginal distributions. For the former, we introduce a term named maximum in-domain likelihood to maintain the discrimination of the learned domain-invariant representation space. For the latter, we approximate the ground-truth marginal distribution with source domains under a reasonable convex hull assumption. Finally, a constrained maximum cross-domain likelihood (CMCL) optimization problem is deduced, by solving which the joint distributions are naturally aligned. An alternating optimization strategy is carefully designed to approximately solve this optimization problem. Extensive experiments on four standard benchmark datasets, i.e., Digits-DG, PACS, Office-Home, and miniDomainNet, highlight the superior performance of our method.

Source-Free Active Domain Adaptation via Augmentation-Based Sample Query and Progressive Model Adaptation

Active domain adaptation (ADA), which enormously improves the performance of unsupervised domain adaptation (UDA) at the expense of annotating limited target data, has attracted a surge of interest. However, in real-world applications, the source data in conventional ADA are not always accessible due to data privacy and security issues. To alleviate this dilemma, we introduce a more practical and challenging setting, dubbed as source-free ADA (SFADA), where one can select a small quota of target samples for label query to assist the model learning, but labeled source data are unavailable. Therefore, how to query the most informative target samples and mitigate the domain gap without the aid of source data are two key challenges in SFADA. To address SFADA, we propose a unified method SQAdapt via augmentation-based ample uery and progressive model Adapt ation. In specific, an active selection module (ASM) is built for target label query, which exploits data augmentation to select the most informative target samples with high predictive sensitivity and uncertainty. Then, we further introduce a classifier adaptation module (CAM) to leverage both the labeled and unlabeled target data for progressively calibrating the classifier weights. Meanwhile, the source-like target samples with low selection scores are taken as source surrogates to realize the distribution alignment in the source-free scenario by the proposed distribution alignment module (DAM). Moreover, as a general active label query method, SQAdapt can be easily integrated into other source-free UDA (SFUDA) methods, and improve their performance. Comprehensive experiments on multiple benchmarks have shown that SQAdapt can achieve superior performance and even surpass most of the ADA methods.

Domain Adaptation via Prompt Learning

Unsupervised domain adaptation (UDA) aims to adapt models learned from a well-annotated source domain to a target domain, where only unlabeled samples are given. Current UDA approaches learn domain-invariant features by aligning source and target feature spaces through statistical discrepancy minimization or adversarial training. However, these constraints could lead to the distortion of semantic feature structures and loss of class discriminability. In this article, we introduce a novel prompt learning paradigm for UDA, named domain adaptation via prompt learning (DAPrompt). In contrast to prior works, our approach learns the underlying label distribution for target domain rather than aligning domains. The main idea is to embed domain information into prompts, a form of representation generated from natural language, which is then used to perform classification. This domain information is shared only by images from the same domain, thereby dynamically adapting the classifier according to each domain. By adopting this paradigm, we show that our model not only outperforms previous methods on several cross-domain benchmarks but also is very efficient to train and easy to implement.

Prompt-Driven Latent Domain Generalization for Medical Image Classification

Deep learning models for medical image analysis easily suffer from distribution shifts caused by dataset artifact bias, camera variations, differences in the imaging station, etc., leading to unreliable diagnoses in real-world clinical settings. Domain generalization (DG) methods, which aim to train models on multiple domains to perform well on unseen domains, offer a promising direction to solve the problem. However, existing DG methods assume domain labels of each image are available and accurate, which is typically feasible for only a limited number of medical datasets. To address these challenges, we propose a unified DG framework for medical image classification without relying on domain labels, called Prompt-driven Latent Domain Generalization (PLDG). PLDG consists of unsupervised domain discovery and prompt learning. This framework first discovers pseudo domain labels by clustering the bias-associated style features, then leverages collaborative domain prompts to guide a Vision Transformer to learn knowledge from discovered diverse domains. To facilitate cross-domain knowledge learning between different prompts, we introduce a domain prompt generator that enables knowledge sharing between domain prompts and a shared prompt. A domain mixup strategy is additionally employed for more flexible decision margins and mitigates the risk of incorrect domain assignments. Extensive experiments on three medical image classification tasks and one debiasing task demonstrate that our method can achieve comparable or even superior performance than conventional DG algorithms without relying on domain labels. Our code is publicly available at https://github.com/SiyuanYan1/PLDG/tree/main.

Feature decoupling integrated domain generalization network for bearing fault diagnosis under unknown operating conditions

In real engineering scenarios, the complex and variable operating conditions of mechanical equipment lead to distributional differences between the collected fault data and the training data. This distribution difference can lead to the failure of deep learning-based diagnostic models. Extracting generalized diagnostic knowledge from the source domain in scenarios where the target domain is not visible is the key to solving this problem. To this end, in this paper, we propose a domain generalization network for diagnosing bearing faults under unknown operating conditions, i.e., Feature Decoupled Integrated Domain Generalization Network (FDIDG). First, we propose a "feature decoupling" algorithm to uncover generalized representations of fault features from multiple source domains. The algorithm aims to explore the generalized representations of fault features by shrinking the distribution of data from multiple source domains and further generalize the fault features to unknown domains to reduce the coupling between fault features and operating conditions. Second, the diagnostic accuracy of the model under unknown operating conditions is further improved by adopting a multi-expert integration strategy in the decision-making stage and utilizing domain-private features to reduce the negative impact of edge samples on diagnosis. We conducted several sets of cross-domain experiments on both public and private datasets, and the results show that FDIDG has excellent generalization capabilities.

Source-free unsupervised domain adaptation: A survey

Unsupervised domain adaptation (UDA) via deep learning has attracted appealing attention for tackling domain-shift problems caused by distribution discrepancy across different domains. Existing UDA approaches highly depend on the accessibility of source domain data, which is usually limited in practical scenarios due to privacy protection, data storage and transmission cost, and computation burden. To tackle this issue, many source-free unsupervised domain adaptation (SFUDA) methods have been proposed recently, which perform knowledge transfer from a pre-trained source model to the unlabeled target domain with source data inaccessible. A comprehensive review of these works on SFUDA is of great significance. In this paper, we provide a timely and systematic literature review of existing SFUDA approaches from a technical perspective. Specifically, we categorize current SFUDA studies into two groups, i.e., white-box SFUDA and black-box SFUDA, and further divide them into finer subcategories based on different learning strategies they use. We also investigate the challenges of methods in each subcategory, discuss the advantages/disadvantages of white-box and black-box SFUDA methods, conclude the commonly used benchmark datasets, and summarize the popular techniques for improved generalizability of models learned without using source data. We finally discuss several promising future directions in this field.

Confounder balancing in adversarial domain adaptation for pre-trained large models fine-tuning

The excellent generalization, contextual learning, and emergence abilities in the pre-trained large models (PLMs) handle specific tasks without direct training data, making them the better foundation models in the adversarial domain adaptation (ADA) methods to transfer knowledge learned from the source domain to target domains. However, existing ADA methods fail to account for the confounder properly, which is the root cause of the source data distribution that differs from the target domains. This study proposes a confounder balancing method in adversarial domain adaptation for PLMs fine-tuning (CadaFT), which includes a PLM as the foundation model for a feature extractor, a domain classifier and a confounder classifier, and they are jointly trained with an adversarial loss. This loss is designed to improve the domain-invariant representation learning by diluting the discrimination in the domain classifier. At the same time, the adversarial loss also balances the confounder distribution among source and unmeasured domains in training. Compared to newest ADA methods, CadaFT can correctly identify confounders in domain-invariant features, thereby eliminating the confounder biases in the extracted features from PLMs. The confounder classifier in CadaFT is designed as a plug-and-play and can be applied in the confounder measurable, unmeasurable, or partially measurable environments. Empirical results on natural language processing and computer vision downstream tasks show that CadaFT outperforms the newest GPT-4, LLaMA2, ViT and ADA methods.

Prediction of Cytochrome P450 Inhibition Using a Deep Learning Approach and Substructure Pattern Recognition

Cytochrome P450 (CYP) is a family of enzymes that are responsible for about 75% of all metabolic reactions. Among them, CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4 participate in the metabolism of most drugs and mediate many adverse drug reactions. Therefore, it is necessary to estimate the chemical inhibition of Cytochrome P450 enzymes in drug discovery and the food industry. In the past few decades, many computational models have been reported, and some provided good performance. However, there are still several issues that should be resolved for these models, such as single isoform, models with unbalanced performance, lack of structural characteristics analysis, and poor availability. In the present study, the deep learning models based on python using the Keras framework and TensorFlow were developed for the chemical inhibition of each CYP isoform. These models were established based on a large data set containing 85715 compounds extracted from the PubChem bioassay database. On external validation, the models provided good AUC values with 0.97, 0.94, 0.94, 0.96, and 0.94 for CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4, respectively. The models can be freely accessed on the Web server named CYPi-DNNpredictor (cypi.sapredictor.cn), and the codes for the model were made open source in the Supporting Information. In addition, we also analyzed the structural characteristics of chemicals with CYP450 inhibition and detected the structural alerts (SAs), which should be responsible for the inhibition. The SAs were also made available online, named CYPi-SAdetector (cypisa.sapredictor.cn). The models can be used as a powerful tool for the prediction of CYP450 inhibitors, and the SAs should provide useful information for the mechanisms of Cytochrome P450 inhibition.

It takes two: Dual Branch Augmentation Module for domain generalization

Although great success has been achieved in various computer vision tasks, deep neural networks (DNNs) suffer dramatic performance degradation when evaluated on out-of-distribution data. Domain generalization (DG) is proposed to handle this problem by learning domain-agnostic information from multiple source domains to generalize well on unseen target domains. Several methods resort to Fourier transform due to its simplicity and efficiency. They argue that amplitude spectra imply domain-specific information, which should be suppressed, while phase counterparts imply domain-agnostic information, which should be preserved. However, these methods only suppress the domain-specific information in source domains and neglect the relationship with target domains, leading to the persistence of the domain gap. Besides, these methods preserve domain-agnostic information by keeping phase components unchanged, causing them to be underutilized. In this paper, we propose Dual Branch Augmentation Module (DBAM) by leveraging Fourier transform and taking advantage of both amplitude and phase spectra. For the amplitude branch, we propose Inner-domain Amplitude Distribution Rectification (IADR) and Cross-domain Amplitude Dirichlet Mixup (CADM) to stabilize the training process and explore more feature space. In addition, we propose Test-time Amplitude Prototype Calibration (TAPC) to construct the connection between source and target domains during evaluation to further mitigate the domain gap. For the phase branch, we propose Random Symmetric Phase Perturbation (RSPP) to enhance the robustness for recognizing domain-agnostic information. With the combined contributions of the two branches, DBAM significantly surpasses other state-of-the-art (SOTA) methods. Extensive experiments on four benchmarks and further analysis demonstrate the effectiveness of DBAM.

Bearing fault detection by using graph autoencoder and ensemble learning

The research and application of bearing fault diagnosis techniques are crucial for enhancing equipment reliability, extending bearing lifespan, and reducing maintenance expenses. Nevertheless, most existing methods encounter challenges in discriminating between signals from machines operating under normal and faulty conditions, leading to unstable detection results. To tackle this issue, the present study proposes a novel approach for bearing fault detection based on graph neural networks and ensemble learning. Our key contribution is a novel stochasticity-based compositional method that transforms Euclidean-structured data into a graph format for processing by graph neural networks, with feature fusion and a newly proposed ensemble learning strategy for outlier detection specifically designed for bearing fault diagnosis. This approach marks a significant advancement in accurately identifying bearing faults, highlighting our study's pivotal role in enhancing diagnostic methodologies.

Enhancing bank marketing strategies with ensemble learning: Empirical analysis

In order to enhance market share and competitiveness, large banks are increasingly focusing on promoting marketing strategies. However, the traditional bank marketing strategy often leads to the homogenization of customer demand, making it challenging to distinguish among various products. To address this issue, this paper presents a customer demand learning model based on financial datasets and optimizes the distribution model of bank big data channels through induction to rectify the imbalance in bank customer transaction data. By comparing the prediction models of random forest model and support vector machine (SVM), this paper analyzes the ability of the prediction model based on ensemble learning to significantly enhance the market segmentation of e-commerce banks. The empirical results reveal that the accuracy of random forest model reaches 92%, while the accuracy of SVM model reaches 87%. This indicates that the ensemble learning model has higher accuracy and forecasting ability than the single model. It enables the bank marketing system to implement targeted marketing, effectively maintain the relationship between customers and banks, and significantly improve the success probability of product marketing. Meanwhile, the marketing model based on ensemble learning has achieved a sales growth rate of 20% and improved customer satisfaction by 30%. This demonstrates that the implementation of the ensemble learning model has also significantly elevated the overall marketing level of bank e-commerce services. Therefore, this paper offers valuable academic guidance for bank marketing decision-making and holds important academic and practical significance in predicting bank customer demand and optimizing product marketing strategy.

Contradiction in text review and apps rating: prediction using textual features and transfer learning

Mobile app stores, such as Google Play, have become famous platforms for practically all types of software and services for mobile phone users. Users may browse and download apps via app stores, which also help developers monitor their apps by allowing users to rate and review them. App reviews may contain the user's experience, bug details, requests for additional features, or a textual rating of the app. These ratings can be frequently biased due to inadequate votes. However, there are significant discrepancies between the numerical ratings and the user reviews. This study uses a transfer learning approach to predict the numerical ratings of Google apps. It benefits from user-provided numeric ratings of apps as the training data and provides authentic ratings of mobile apps by analyzing users' reviews. A transfer learning-based model ELMo is proposed for this purpose which is based on the word vector feature representation technique. The performance of the proposed model is compared with three other transfer learning and five machine learning models. The dataset is scrapped from the Google Play store which extracts the data from 14 different categories of apps. First, biased and unbiased user rating is segregated using TextBlob analysis to formulate the ground truth, and then classifiers prediction accuracy is evaluated. Results demonstrate that the ELMo classifier has a high potential to predict authentic numeric ratings with user actual reviews.

Weighted Domain Adaptation Using the Graph-Structured Dataset Representation for Machinery Fault Diagnosis under Varying Operating Conditions

Data-driven fault diagnosis has received significant attention in the era of big data. Most data-driven methods have been developed under the assumption that both training and test data come from identical data distributions. However, in real-world industrial scenarios, data distribution often changes due to varying operating conditions, leading to a degradation of diagnostic performance. Although several domain adaptation methods have shown their feasibility, existing methods have overlooked metadata from the manufacturing process and treated all domains uniformly. To address these limitations, this article proposes a weighted domain adaptation method using a graph-structured dataset representation. Our framework involves encoding a collection of datasets into the proposed graph structure, which captures relations between datasets based on metadata and raw data simultaneously. Then, transferability scores of candidate source datasets for a target are estimated using the constructed graph and a graph embedding model. Finally, the fault diagnosis model is established with a voting ensemble of the base classifiers trained on candidate source datasets and their estimated transferability scores. For validation, two case studies on rotor machinery, specifically tool wear and bearing fault detection, were conducted. The experimental results demonstrate the effectiveness and superiority of the proposed method over other existing domain adaptation methods.

Latent Feature Disentanglement for Visual Domain Generalization

Despite remarkable success in a variety of computer vision applications, it is well-known that deep learning can fail catastrophically when presented with out-of-distribution data, where there are usually style differences between the training and test images. Toward addressing this challenge, we consider the domain generalization problem, wherein predictors are trained using data drawn from a family of related training (source) domains and then evaluated on a distinct and unseen test domain. Naively training a model on the aggregate set of data (pooled from all source domains) has been shown to perform suboptimally, since the information learned by that model might be domain-specific and generalizes imperfectly to test domains. Data augmentation has been shown to be an effective approach to overcome this problem. However, its application has been limited to enforcing invariance to simple transformations like rotation, brightness change, etc. Such perturbations do not necessarily cover plausible real-world variations that preserve the semantics of the input (such as a change in the image style). In this paper, taking the advantage of multiple source domains, we propose a novel approach to express and formalize robustness to these kind of real-world image perturbations. The three key ideas underlying our formulation are (1) leveraging disentangled representations of the images to define different factors of variations, (2) generating perturbed images by changing such factors composing the representations of the images, (3) enforcing the learner (classifier) to be invariant to such changes in the images. We use image-to-image translation models to demonstrate the efficacy of this approach. Based on this, we propose a domain-invariant regularization (DIR) loss function that enforces invariant prediction of targets (class labels) across domains which yields improved generalization performance. We demonstrate the effectiveness of our approach on several widely used datasets for the domain generalization problem, on all of which our results are competitive with the state-of-the-art.

Early Diagnosis and Classification of Fetal Health Status from a Fetal Cardiotocography Dataset Using Ensemble Learning

(1) Background: According to the World Health Organization (WHO), 6.3 million intrauterine fetal deaths occur every year. The most common method of diagnosing perinatal death and taking early precautions for maternal and fetal health is a nonstress test (NST). Data on the fetal heart rate and uterus contractions from an NST device are interpreted based on a trace printer's output, allowing for a diagnosis of fetal health to be made by an expert. (2) Methods: in this study, a predictive method based on ensemble learning is proposed for the classification of fetal health (normal, suspicious, pathology) using a cardiotocography dataset of fetal movements and fetal heart rate acceleration from NST tests. (3) Results: the proposed predictor achieved an accuracy level above 99.5% on the test dataset. (4) Conclusions: from the experimental results, it was observed that a fetal health diagnosis can be made during NST using machine learning.

Feature Alignment by Uncertainty and Self-Training for Source-Free Unsupervised Domain Adaptation

Most unsupervised domain adaptation (UDA) methods assume that labeled source images are available during model adaptation. However, this assumption is often infeasible owing to confidentiality issues or memory constraints on mobile devices. Some recently developed approaches do not require source images during adaptation, but they show limited performance on perturbed images. To address these problems, we propose a novel source-free UDA method that uses only a pre-trained source model and unlabeled target images. Our method captures the aleatoric uncertainty by incorporating data augmentation and trains the feature generator with two consistency objectives. The feature generator is encouraged to learn consistent visual features away from the decision boundaries of the head classifier. Thus, the adapted model becomes more robust to image perturbations. Inspired by self-supervised learning, our method promotes inter-space alignment between the prediction space and the feature space while incorporating intra-space consistency within the feature space to reduce the domain gap between the source and target domains. We also consider epistemic uncertainty to boost the model adaptation performance. Extensive experiments on popular UDA benchmark datasets demonstrate that the proposed source-free method is comparable or even superior to vanilla UDA methods. Moreover, the adapted models show more robust results when input images are perturbed.

Generative Inference Network for Imbalanced Domain Generalization

Domain generalization (DG) aims to learn transferable knowledge from multiple source domains and generalize it to the unseen target domain. To achieve such expectation, the intuitive solution is to seek domain-invariant representations via generative adversarial mechanism or minimization of cross-domain discrepancy. However, the widespread imbalanced data scale problem across source domains and category in real-world applications becomes the key bottleneck of improving generalization ability of model due to its negative effect on learning the robust classification model. Motivated by this observation, we first formulate a practical and challenging imbalance domain generalization (IDG) scenario, and then propose a straightforward but effective novel method generative inference network (GINet), which augments reliable samples for minority domain/category to promote discriminative ability of the learned model. Concretely, GINet utilizes the available cross-domain images from the identical category and estimates their common latent variable, which derives to discover domain-invariant knowledge for unseen target domain. According to these latent variables, our GINet further generates more novel samples with optimal transport constraint and deploys them to enhance the desired model with more robustness and generalization ability. Considerable empirical analysis and ablation studies on three popular benchmarks under normal DG and IDG setups suggests the advantage of our method over other DG methods on elevating model generalization. The source code is available in GitHub https://github.com/HaifengXia/IDG.

Memorizing Complementation Network for Few-Shot Class-Incremental Learning

Few-shot Class-Incremental Learning (FSCIL) aims at learning new concepts continually with only a few samples, which is prone to suffer the catastrophic forgetting and overfitting problems. The inaccessibility of old classes and the scarcity of the novel samples make it formidable to realize the trade-off between retaining old knowledge and learning novel concepts. Inspired by that different models memorize different knowledge when learning novel concepts, we propose a Memorizing Complementation Network (MCNet) to ensemble multiple models that complements the different memorized knowledge with each other in novel tasks. Additionally, to update the model with few novel samples, we develop a Prototype Smoothing Hard-mining Triplet (PSHT) loss to push the novel samples away from not only each other in current task but also the old distribution. Extensive experiments on three benchmark datasets, e.g., CIFAR100, miniImageNet and CUB200, have demonstrated the superiority of our proposed method.

Style Uncertainty Based Self-Paced Meta Learning for Generalizable Person Re-Identification

Domain generalizable person re-identification (DG ReID) is a challenging problem, because the trained model is often not generalizable to unseen target domains with different distribution from the source training domains. Data augmentation has been verified to be beneficial for better exploiting the source data to improve the model generalization. However, existing approaches primarily rely on pixel-level image generation that requires designing and training an extra generation network, which is extremely complex and provides limited diversity of augmented data. In this paper, we propose a simple yet effective feature based augmentation technique, named Style-uncertainty Augmentation (SuA). The main idea of SuA is to randomize the style of training data by perturbing the instance style with Gaussian noise during training process to increase the training domain diversity. And to better generalize knowledge across these augmented domains, we propose a progressive learning to learn strategy named Self-paced Meta Learning (SpML) that extends the conventional one-stage meta learning to multi-stage training process. The rationality is to gradually improve the model generalization ability to unseen target domains by simulating the mechanism of human learning. Furthermore, conventional person Re-ID loss functions are unable to leverage the valuable domain information to improve the model generalization. So we further propose a distance-graph alignment loss that aligns the feature relationship distribution among domains to facilitate the network to explore domain-invariant representations of images. Extensive experiments on four large-scale benchmarks demonstrate that our SuA-SpML achieves state-of-the-art generalization to unseen domains for person ReID.

Contrastive-ACE: Domain Generalization Through Alignment of Causal Mechanisms

Domain generalization aims to learn knowledge invariant across different distributions while semantically meaningful for downstream tasks from multiple source domains, to improve the model's generalization ability on unseen target domains. The fundamental objective is to understand the underlying "invariance" behind these observational distributions and such invariance has been shown to have a close connection to causality. While many existing approaches make use of the property that causal features are invariant across domains, we consider the invariance of the average causal effect of the features to the labels. This invariance regularizes our training approach in which interventions are performed on features to enforce stability of the causal prediction by the classifier across domains. Our work thus sheds some light on the domain generalization problem by introducing invariance of the mechanisms into the learning process. Experiments on several benchmark datasets demonstrate the performance of the proposed method against SOTAs. The codes are available at: https://github.com/lithostark/Contrastive-ACE.

Speech Enhancement Model Synthesis Based on Federal Learning for Industrial CPS in Multiple Noise Conditions

Real-time acquisition of industrial production data and rapid response to changes in the external environment are key to ensuring the symmetry of a CPS. However, during industrial production, the collected data are inevitably disturbed by environmental noise, which has a huge impact on the subsequent data processing of a CPS. The types of noise vary greatly in different work scenarios in a factory. Meanwhile, barriers such as data privacy protection and copyright restrictions create great difficulties for model synthesis in the information space. A speech enhancement model with teacher–student architecture based on federal knowledge distillation is proposed to alleviate this problem. (1) We pre-train teacher models under different noise conditions to create multiple teacher models with symmetry and excelling in the suppression of a priori noise. (2) We construct a symmetric model–student model of the physical space of the teacher model trained on public data and transfer the knowledge of the teacher model to the student model. The student model can suppress multiple types of noise. Notably, with the TIMIT dataset and the NoiseX92 noise set, the accuracy of the proposed method improved by an average of 1.00% over the randomly specified teacher method in the PESQ metric and 0.17% for STOI.

Dynamic Instance Domain Adaptation

Most existing studies on unsupervised domain adaptation (UDA) assume that each domain's training samples come with domain labels (e.g., painting, photo). Samples from each domain are assumed to follow the same distribution and the domain labels are exploited to learn domain-invariant features via feature alignment. However, such an assumption often does not hold true-there often exist numerous finer-grained domains (e.g., dozens of modern painting styles have been developed, each differing dramatically from those of the classic styles). Therefore, forcing feature distribution alignment across each artificially-defined and coarse-grained domain can be ineffective. In this paper, we address both single-source and multi-source UDA from a completely different perspective, which is to view each instance as a fine domain. Feature alignment across domains is thus redundant. Instead, we propose to perform dynamic instance domain adaptation (DIDA). Concretely, a dynamic neural network with adaptive convolutional kernels is developed to generate instance-adaptive residuals to adapt domain-agnostic deep features to each individual instance. This enables a shared classifier to be applied to both source and target domain data without relying on any domain annotation. Further, instead of imposing intricate feature alignment losses, we adopt a simple semi-supervised learning paradigm using only a cross-entropy loss for both labeled source and pseudo labeled target data. Our model, dubbed DIDA-Net, achieves state-of-the-art performance on several commonly used single-source and multi-source UDA datasets including Digits, Office-Home, DomainNet, Digit-Five, and PACS.