An Improved Method for Cross-Domain Pedestrian Re-identification

Domain-transfer learning is a machine learning task to explore a source domain data set to help the learning problem in a target domain. Usually, the source domain has sufficient labeled data, while the target domain does not. In this paper, we propose a novel domain-transfer convolutional model by mapping a target domain data sample to a proxy in the source domain and applying a source domain model to the proxy for the purpose of prediction. In our framework, we firstly represent both source and target domains to feature vectors by two convolutional neural networks and then construct a proxy for each target domain sample in the source domain space. The proxy is supposed to be matched to the corresponding target domain sample convolutional representation vector well. To measure the matching quality, we proposed to maximize their squared-loss mutual information (SMI) between the proxy and target domain samples. We further develop a novel neural SMI estimator based on a parametric density ratio estimation function. Moreover, we also propose to minimize the classification error of both source domain samples and target domain proxies. The classification responses are also smoothened by manifolds of both the source domain and proxy space. By minimizing an objective function of SMI, classification error, and manifold regularization, we learn the convolutional networks of both source and target domains. In this way, the proxy of a target domain sample can be matched to the source domain data and thus benefits from the rich supervision information of the source domain. We design an iterative algorithm to update the parameters alternately and test it over benchmark data sets of abnormal behavior detection in video, Amazon product reviews sentiment analysis, etc.

Recent progress in machine learning has been mainly due to the availability of large amounts of annotated data used for training complex models with deep architectures. Annotating this training data becomes burdensome and creates a major bottleneck in maintaining machine-learning databases. Moreover, these trained models fail to generalize to new categories or new varieties of the same categories. This is because new categories or new varieties have data distribution different from the training data distribution. To tackle these problems, this thesis proposes to develop a family of transfer-learning techniques that can deal with different training (source) and testing (target) distributions with the assumption that the availability of annotated data is limited in the testing domain. This is done by using the auxiliary data-abundant source domain from which useful knowledge is transferred that can be applied to data-scarce target domain. This transferable knowledge serves as a prior that biases target-domain predictions and prevents the target-domain model from overfitting. Specifically, we explore structural priors that encode relational knowledge between different data entities, which provides more informative bias than traditional priors. The choice of the structural prior depends on the information availability and the similarity between the two domains. Depending on the domain similarity and the information availability, we divide the transfer learning problem into four major categories and propose different structural priors to solve each of these sub-problems. This thesis first focuses on the unsupervised-domain-adaptation problem, where we propose to minimize domain discrepancy by transforming labeled source-domain data to be close to unlabeled target-domain data. For this problem, the categories remain the same across the two domains and hence we assume that the structural relationship between the source-domain samples is carried over to the target domain. Thus, graph or hyper-graph is constructed as the structural prior from both domains and a graph/hyper-graph matching formulation is used to transform samples in the source domain to be closer to samples in the target domain. An efficient optimization scheme is then proposed to tackle the time and memory inefficiencies associated with the matching problem. The few-shot learning problem is studied next, where we propose to transfer knowledge from source-domain categories containing abundantly labeled data to novel categories in the target domain that contains only few labeled data. The knowledge transfer biases the novel category predictions and prevents the model from overfitting. The knowledge is encoded using a neural-network-based prior that transforms a data sample to its corresponding class prototype. This neural network is trained from the source-domain data and applied to the target-domain data, where it transforms the few-shot samples to the novel-class prototypes for better recognition performance. The few-shot learning problem is then extended to the situation, where we do not have access to the source-domain data but only have access to the source-domain class prototypes. In this limited information setting, parametric neural-network-based priors would overfit to the source-class prototypes and hence we seek a non-parametric-based prior using manifolds. A piecewise linear manifold is used as a structural prior to fit the source-domain-class prototypes. This structure is extended to the target domain, where the novel-class prototypes are found by projecting the few-shot samples onto the manifold. Finally, the zero-shot learning problem is addressed, which is an extreme case of the few-shot learning problem where we do not have any labeled data in the target domain. However, we have high-level information for both the source and target domain categories in the form of semantic descriptors. We learn the relation between the sample space and the semantic space, using a regularized neural network so that classification of the novel categories can be carried out in a common representation space. This same neural network is then used in the target domain to relate the two spaces. In case we want to generate data for the novel categories in the target domain, we can use a constrained generative adversarial network instead of a traditional neural network. Thus, we use structural priors like graphs, neural networks and manifolds to relate various data entities like samples, prototypes and semantics for these different transfer learning sub-problems. We explore additional post-processing steps like pseudo-labeling, domain adaptation and calibration and enforce algorithmic and architectural constraints to further improve recognition performance. Experimental results on standard transfer learning image recognition datasets produced competitive results with respect to previous work. Further experimentation and analyses of these methods provided better understanding of machine learning as well.

Reliable prediction of assembly precision is important for quality control of customized mechanical products characterized by individual customization, small batch size, and multiple varieties, resulting in insufficient samples for predicting assembly performance. A customized mechanical product assembly precision prediction method based on generative adversarial networks and feature transfer learning (GAN-FTL) is proposed in this paper. A GAN is built based on high quality data (source domain) to generate auxiliary samples with high fidelity and large sample size. A support vector machine is used to generate pseudo-tags for auxiliary samples. Features of source domain, target domain and auxiliary samples from different distributions are transferred to the same distribution to achieve multi-source fusion of measured and simulated data using FTL. Data after FTL is used to train the assembly precision prediction model. The elevator guide rail assembly is taken as the case study. T70/B and T90/B guide rail assembly are selected as the source and target domains, respectively. FTL was performed between the source and target domains, with different sample sets for comparison and compared with five different methods. Experimental results show that the prediction accuracy of the target domain is improved when the auxiliary sample size is 300, 400, and 500, and the accuracy improvement of the five methods are 15.37%, 12.17%, 9.68%, 6.29%, and 4.31%, respectively, which verified the effectiveness and usability of the proposed assembly precision prediction method based on GAN-FTL.

Importance-weighted conditional adversarial network for unsupervised domain adaptation

Domain Adaptation (DA) methods are usually carried out by means of simply reducing the marginal distribution differences between the source and target domains, and subsequently using the resultant trained classifier, namely source classifier, for use in the target domain. However, in many cases, the true predictive distributions of the source and target domains can be vastly different especially when their class distributions are skewed, causing the issues of sample selection bias in DA. Hence, DA methods which leverage the source labeled data may suffer from poor generalization in the target domain, resulting in negative transfer. In addition, we observed that many DA methods use either a source classifier or a linear combination of source classifiers with a fixed weighting for predicting the target unlabeled data. Essentially, the labels of the target unlabeled data are spanned by the prediction of these source classifiers. Motivated by these observations, in this paper, we propose to construct many source classifiers of diverse biases and learn the weight for each source classifier by directly minimizing the structural risk defined on the target unlabeled data so as to heal the possible sample selection bias. Since the weights are learned by maximizing the margin of separation between opposite classes on the target unlabeled data, the proposed method is established here as Maximal Margin Target Label Learning (MMTLL), which is in a form of Multiple Kernel Learning problem with many label kernels. Extensive experimental studies of MMTLL against several state-of-the-art methods on the Sentiment and Newsgroups datasets with various imbalanced class settings showed that MMTLL exhibited robust accuracies on all the settings considered and was resilient to negative transfer, in contrast to other counterpart methods which suffered significantly in prediction accuracy.

In the absence of the labeled samples in a domain referred to as target domain, Domain Adaptation (DA) techniques come in handy. Generally, DA techniques assume there are available source domains that share similar predictive function with the target domain. Two core challenges of DA typically arise, variance that exists between source and target domains, and the inherent source hypothesis bias. In this paper, we first propose a Stability Transfer criterion for selecting relevant source domains with low variance. With this criterion, we introduce a TARget learning Assisted by Source Classifier Adaptation (TARASCA) method to address the two core challenges that have impeded the performances of DA techniques. To verify the robustness of TARASCA, extensive experimental studies are carried out with comparison to several state-of-the-art DA methods on the real-world Sentiment and Newsgroups datasets, where various settings for the class ratios of the source and target domains are considered.

Bearing is an important component in mechanical equipment. Its main function is to support the rotating mechanical body and reduce the friction coefficient and axial load. In the actual operating environment, the bearings are affected by complex working conditions and other factors. Therefore, it is very difficult to effectively obtain data that meets the conditions of independent and identical distribution of training data and test data, which result in unsatisfactory fault diagnosis results. As a transfer learning method, joint distribution adaptive (JDA) can effectively solve the learning problem of inconsistent distribution of training data and test data. In this paper, a new bearing fault diagnosis method based on JDA and deep belief network (DBN) with improved sparrow search algorithm (CWTSSA), namely JACADN is proposed. In the JACADN, the JDA is employed to carry out feature transfer between the source domain samples and target domain samples, that is, the source domain samples and target domain samples are mapped into the same feature space by the kernel function. Then the maximum mean difference is used as the metric to reduce the joint distribution difference between the samples in the two domains. Aiming at the parameter selection of the DBN, an improved sparrow search algorithm (CWTSSA) with global optimization ability is used to optimize the parameters of the DBN in order to construct an optimized DBN model. The obtained source domain samples and target domain samples are divided into training set and test set, which are input the optimized DBN to construct a bearing fault diagnosis model for improving the diagnosis accuracy. The effectiveness of the proposed method is verified by vibration data of QPZZ-II rotating machinery. The experimental results show that the proposed JACADN method can effectively improve the fault diagnosis accuracy of rolling bearings under variable operating conditions.

In many real-world applications, the domain of model learning (referred as source domain) is usually inconsistent with or even different from the domain of testing (referred as target domain), which makes the learnt model degenerate in target domain, i.e., the test domain. To alleviate the discrepancy between source and target domains, we propose a domain adaptation method, named as Bi-shifting Auto-Encoder network (BAE). The proposed BAE attempts to shift source domain samples to target domain, and also shift the target domain samples to source domain. The non-linear transformation of BAE ensures the feasibility of shifting between domains, and the distribution consistency between the shifted domain and the desirable domain is constrained by sparse reconstruction between them. As a result, the shifted source domain is supervised and follows similar distribution as target domain. Therefore, any supervised method can be applied on the shifted source domain to train a classifier for classification in target domain. The proposed method is evaluated on three domain adaptation scenarios of face recognition, i.e., domain adaptation across view angle, ethnicity, and imaging sensor, and the promising results demonstrate that our proposed BAE can shift samples between domains and thus effectively deal with the domain discrepancy.

Dynamic source domain selection: An adaptive EEG transfer learning framework for mitigating negative transfer.

An intelligent fault diagnosis model based on deep neural network for few-shot fault diagnosis

In this paper, we propose a new unsupervised domain adaptation method for document classification. We address the problem of domain adaptation for document classification where the source and target domains do not differ significantly and there is no labeled data in the target domain. In this case, we can use conventional semi-supervised learning. Thus, we use the naive Bayes-based expectation-maximization method (NBEM) which is very effective for document classification. However, NBEM does not utilize the difference between a source domain and a target domain. We combine NBEM with the feature weighted method for domain adaptation, referred to as “self-training feature weight” (STFW). Our proposed method alternately uses NBEM and STFW to gradually improve document classification precision for a target domain. This method significantly outperforms the conventional unsupervised methods for domain adaptation.

In real-world hyperspectral image (HSI) classification, the limited annotated examples usually lead to the insufficiently trained classifier, which further generates low classification accuracy. To overcome this challenge, Domain Adaptation (DA) methods have been developed to transfer learnable knowledge from the external HSIs with sufficient labeled examples ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e</i> ., the source domain) to the interested HSI with scarce labeled examples ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e</i> ., the target domain). Conventional DA approaches often pseudo-label the examples with high classification confidence and then incorporate them in the training process. However, due to the significant domain gap, relying solely on the confident examples may not be adequate to achieve satisfactory performance. Therefore, this paper proposes an Interactive Easy-to-Hard Domain Adaptation method (IEH-DA) to arrange the adaptation process so that the “easy” examples are adapted ahead of the “hard” ones. In an early stage, the easy examples with high pseudo-labeling confidence are selected for the adversarial learning based DA. In a later stage, the “hard” examples with high informativity are further selected, and they are interactively labeled by human expert to provide accurate supervision information for adaptation. As a result, the examples in target domain are used in an easy-to-hard way, which forms a curriculum sequence for orderly model training. Extensive experiments conducted on typical public datasets demonstrate that IEH-DA outperforms other state-of-the-art DA methods for HSI classification.

Domain adaptation is one of the prominent strategies for handling both domain shift, that is widely encountered in large-scale land use/land cover map calculation, and the scarcity of pixel-level ground truth that is crucial for supervised semantic segmentation. Studies focusing on adversarial domain adaptation via re-styling source domain samples, commonly through generative adversarial networks, have reported varying levels of success, yet they suffer from semantic inconsistencies, visual corruptions, and often require a large number of target domain samples. In this letter, we propose a new unsupervised domain adaptation method for the semantic segmentation of very high resolution images, that i) leads to semantically consistent and noise-free images, ii) operates with a single target domain sample (i.e. one-shot) and iii) at a fraction of the number of parameters required from state-of-the-art methods. More specifically an image-to-image translation paradigm is proposed, based on an encoder-decoder principle where latent content representations are mixed across domains, and a perceptual network module and loss function is further introduced to enforce semantic consistency. Cross-city comparative experiments have shown that the proposed method outperforms state-of-the-art domain adaptation methods. Our source code will be available at \url{https://github.com/Sarmadfismael/LRM_I2I}.

Domain adaptation (DA), which leverages labeled data from related source domains, comes in handy when the label information of the target domain is scarce or unavailable. However, as the source data do not come from the same origin as that of the target domain, the predictive distributions of the source and target domains are likely to differ in reality. At the extreme, the predictive distributions of the source domains can differ completely from that of the target domain. In such case, using the learned source classifier to assist in the prediction of target data can result in prediction performance that is poorer than that with the omission of the source data. This phenomenon is established as negative transfer with impact known to be more severe in the multiclass context. To combat negative transfer due to differing predictive distributions across domains, we first introduce the notion of positive transferability for the assessment of synergy between the source and target domains in their prediction models, and we also propose a criterion to measure the positive transferability between sample pairs of different domains in terms of their prediction distributions. With the new measure, a predictive distribution matching (PDM) regularizer and a PDM framework learn the target classifier by favoring source data with large positive transferability while inferring the labels of target unlabeled data. Extensive experiments are conducted to validate the performance efficacy of the proposed PDM framework using several commonly used multidomain benchmark data sets, including Sentiment, Reuters, and Newsgroup, in the context of both binary-class and multiclass domains. Subsequently, the PDM framework is put to work on a real-world scenario pertaining to water cluster molecule identification. The experimental results illustrate the adverse impact of negative transfer on several state-of-the-art DA methods, whereas the proposed framework exhibits excellent and robust predictive performances.

Easy Domain Adaptation for cross-subject multi-view emotion recognition