Unsupervised Deep Domain Adaptation Research Articles

Unsupervised deep domain adaptation algorithm for video based human activity recognition via recurrent neural networks

The integration of cyber-physical systems and artificial intelligent human activity recognition (HAR) applications enables intelligent interactions within a physical environment. In real-world HAR applications, the domain shift between training (source) and testing (target) images captured in different scenarios leads to low classification accuracy. Existing unsupervised domain adaptation (UDA) methods often require some labeled target samples for model adaptation, which limits their practicality. This study proposes a novel unsupervised deep domain adaptation algorithm (UDDAA) for HAR using recurrent neural networks. UDDAA introduces a maximum mean class discrepancy (MMCD) metric, accounting for both inter-class and intra-class differences within each domain. MMCD extends the maximum mean discrepancy to measure the class-level distribution discrepancy across source and target domains, aligning these distributions to enhance domain adaptation performance. Without relying on labeled target data, UDDAA predicts pseudo-labels for the unlabeled target dataset, combining these with labeled source data to train the model on domain-invariant representations. This approach makes UDDAA highly practical for scenarios where labeled target data is difficult or expensive to obtain, enabling human-computer interaction (HCI) systems to function effectively across varied environments and user behaviors. Extensive experiments on benchmark datasets demonstrate UDDAA's superior classification accuracy over existing baselines. Notably, UDDAA achieved 92% and 99% accuracy for University of Central Florida database (UCF) to Human Motion Database (HMDB) and HMDB to UCF transfers, respectively. Additionally, on personal recorded videos with complex backgrounds, it achieved high classification accuracies of 95% for basketball and 90% for football activities, underscoring its generalization ability, robustness, and effectiveness.

S2AC: Self-Supervised Attention Correlation Alignment Based on Mahalanobis Distance for Image Recognition

Susceptibility to domain changes for image classification hinders the application and development of deep neural networks. Domain adaptation (DA) makes use of domain-invariant characteristics to improve the performance of a model trained on labeled data from one domain (source domain) on an unlabeled domain (target) with a different data distribution. But existing DA methods simply use pretrained models (e.g., AlexNet, ResNet) for feature extraction, which are convolutional models that are trapped in localized features and fail to acquire long-distance dependencies. Furthermore, many approaches depend too much on pseudo-labels, which can impair adaptation efficiency and lead to unstable and inconsistent results. In this research, we present S2AC, a novel approach for unsupervised deep domain adaptation, that makes use of a stacked attention architecture as a feature map extractor. Our method can fuse domain discrepancy with minimizing a linear transformation of the second statistics (covariances) extended by the p-norm, while simultaneously designing pretext tasks on heuristics to improve the generality of the learning representation. In addition, we have developed a new trainable relative position embedding that not only reduces the model parameters but also enhances model accuracy and expedites the training process. To illustrate our method’s efficacy and controllability, we designed extensive experiments based on the Office31, Office_Caltech_10, and OfficeHome datasets. To the best of our knowledge, the proposed method is the first attempt at incorporating attention-based networks and self-supervised learning for image domain adaptation, and has shown promising results.

A Compressed Unsupervised Deep Domain Adaptation Model for Efficient Cross-Domain Fault Diagnosis

As one of the most important artificial intelligence-enabled industrial applications, fault diagnosis is vital in the safe, stable and reliable operation of the equipment. Many existing deep learning-based fault diagnosis methods assume that the distribution of training data is the same as that of testing data, which is almost impossible in practical industrial applications. In addition, most of these fault diagnosis methods are generally memory intensive and computationally expensive. A compressed unsupervised deep domain adaption model-based fault diagnosis method is proposed to overcome the above-mentioned two issues. Firstly, a standard unsupervised domain adaption model is designed to extract the features of training data and testing data, respectively. Then, the maximum mean discrepancy term is introduced to minimize the discrepancy between the extracted features of them. Next, the standard model is compressed through iteratively pruning the redundant convolutional channels. Finally, the obtained compressed model is applied to diagnose faults. The performance of the proposed method is verified on the Case Western Reserve University bearing dataset. Experimental results show that the compressed model can significantly reduce the memory occupation, computational cost and inference time compared with the standard model, but still achieve comparable or even better accuracy on ten transfer diagnostic tasks.

Seismic Facies Analysis: A Deep Domain Adaptation Approach

Deep neural networks (DNNs) can learn accurately from large quantities of labeled input data but often fail to do so when labeled data are scarce. DNNs sometimes fail to generalize on test data sampled from different input distributions. Unsupervised deep domain adaptation (DDA) techniques have been proven useful when no labels are available and when distribution shifts are observed in the target domain (TD). In this study, experiments are performed on seismic images of the F3 block 3-D dataset from offshore Netherlands [source domain (SD)] and Penobscot 3-D survey data from Canada (TD). Three geological classes from SD and TD that have similar reflection patterns are considered. A DNN architecture named EarthAdaptNet (EAN) is proposed to semantically segment the seismic images when few classes have data scarcity, and we use a transposed residual unit to replace the traditional dilated convolution in the decoder block. The EAN achieved a pixel-level accuracy >84% and an accuracy of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 70$ </tex-math></inline-formula> % for the minority classes, showing improved performance compared to existing architectures. In addition, we introduce the correlation alignment (CORAL) method to the EAN to create an unsupervised deep domain adaptation network (EAN-DDA) for the classification of seismic reflections from F3 and Penobscot to demonstrate possible approaches when labeled data are unavailable. Maximum class accuracy achieved was <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 99$ </tex-math></inline-formula> % for class 2 of Penobscot with an overall accuracy >50%. Taken together, the EAN-DDA has the potential to classify TD seismic facies classes with high accuracy.

Unsupervised deep domain adaptation framework in remote acoustic time parametric imaging

Intelligent waveform analysis surveying geological structures is a challenging task in remote acoustic measurement for formation detection, which has two problems: (1) time parametric imaging is disturbed by noisy environments and (2) manually annotated data for machine learning are unattainable. These restrict the deployment of advanced parameter extraction methods in imaging instruments. As a potential theory, domain adaptation makes the intelligent prediction implementable in the above situations. Hence, to counter high precision travel time extraction for acoustic imaging, a deep adaptation-picking network (DAPN) framework is proposed, which consists of three modules: signal-to-signal generator, domain adaptation encoder, and feature recognition decoder. The translation module preprocesses the different datasets to improve the training accuracy and confuses the distribution between source and target domains based on the adversarial network. The core convolution neural network achieves travel time measurement, where the backbone extracts content features with modified maximum mean discrepancy embedding. Meanwhile, the decoding structure preserves the signal features in the training process. The effectiveness and feasibility of DAPN are verified by experimental results, suggesting that DAPN accurately extracts time parameters. Compared with traditional parametric imaging schemes, the proposed method has the advantages of high accuracy and anti-noise ability.

A Survey of Unsupervised Deep Domain Adaptation.

Deep learning has produced state-of-the-art results for a variety of tasks. While such approaches for supervised learning have performed well, they assume that training and testing data are drawn from the same distribution, which may not always be the case. As a complement to this challenge, single-source unsupervised domain adaptation can handle situations where a network is trained on labeled data from a source domain and unlabeled data from a related but different target domain with the goal of performing well at test-time on the target domain. Many single-source and typically homogeneous unsupervised deep domain adaptation approaches have thus been developed, combining the powerful, hierarchical representations from deep learning with domain adaptation to reduce reliance on potentially-costly target data labels. This survey will compare these approaches by examining alternative methods, the unique and common elements, results, and theoretical insights. We follow this with a look at application areas and open research directions.

Unsupervised adversarial deep domain adaptation method for potato defects classification

In the last few years, deep learning methods have been proposed to automate the quality control of various agricultural products. Despite the excellent results obtained, one of their main drawbacks is the need for a large annotated dataset to obtain satisfactory performance. Building this dataset is time-consuming and tedious. Moreover, in some real-world applications, vision systems can be modified over time. Such changes may generate a drop in network performance trained with the initial database (source images). To avoid the creation of a new labeled database every time a change in data distribution occurs (target images), several domain adaptation methods have been proposed. In this article, we introduce an unsupervised deep domain adaptation method based on adversarial training. A large dataset is used, including six classes of potatoes: healthy, damaged, greening, black dot, common scab, and black scurf. Two different scenarios of domain adaptation problem are considered. Firstly, a simply modification of the image acquisition system is simulated by artificially increasing the brightness of some white potatoes images (target images). Secondly, a significantly different dataset including red potatoes is introduced. In this setting, white potatoes are used as source images and red tubers as target images. We propose to train a target classifier using a pseudo-label loss, due to the unavailability of target annotations. Experimental results show that a domain adaptation method is mandatory, going from an average F1-score of 0.46 without adaptation, to 0.84 by applying our method. Finally, a comparative analysis is achieved showing that adversarial-based unsupervised domain adaptation methods outperform discrepancy-based approaches.

Xnet: Task-specific attentional domain adaptation for satellite-to-aerial scene

Domain shift remains to be challenging in the current unsupervised deep domain adaptation literature. Although many state-of-the-art models have been proposed recently, non-adversarial and adversarial domain adaptation, as two major types of deep domain adaptation frameworks, cannot fully address this concern. In this paper, our proposed deep domain adaptation model attempts to address this challenge by leveraging the desired structural properties of both non-adversarial and adversarial models. Our contributions are two-fold: We solve the domain shift problem via domain-specific deep feature generator; we further adapt the generated features to the classification task by introducing task-specific attention learning. We name the proposed framework as ‘Xnet’, given the X- shape of the proposed deep neural network. The objective is to demonstrate the effectiveness of the proposed model both on traditional digit recognition tasks and the newly proposed satellite-to-aerial scene adaptation task. The full implementation (based on Tensorflow) as well as the trained networks are available at https://github.com/ubcsip/Xnet.

Adversarial auto‐encoder for unsupervised deep domain adaptation

Unsupervised visual domain adaptation aims to train a classifier that works well on a target domain given labelled source samples and unlabelled target samples. The key issue in unsupervised visual domain adaptation is how to do the feature alignment between source and target domains. Inspired by the adversarial learning in generative adversarial networks, this study proposes a novel adversarial auto-encoder for unsupervised deep domain adaptation. This method incorporates the auto-encoder with the adversarial learning so that the domain similarity and reconstruction information from the decoder can be exploited to facilitate the adversarial domain adaptation in the encoder. Extensive experiments on various visual recognition tasks show that the proposed method performs favourably against or better than competitive state-of-the-art methods.

Joint Domain Alignment and Discriminative Feature Learning for Unsupervised Deep Domain Adaptation

Recently, considerable effort has been devoted to deep domain adaptation in computer vision and machine learning communities. However, most of existing work only concentrates on learning shared feature representation by minimizing the distribution discrepancy across different domains. Due to the fact that all the domain alignment approaches can only reduce, but not remove the domain shift, target domain samples distributed near the edge of the clusters, or far from their corresponding class centers are easily to be misclassified by the hyperplane learned from the source domain. To alleviate this issue, we propose to joint domain alignment and discriminative feature learning, which could benefit both domain alignment and final classification. Specifically, an instance-based discriminative feature learning method and a center-based discriminative feature learning method are proposed, both of which guarantee the domain invariant features with better intra-class compactness and inter-class separability. Extensive experiments show that learning the discriminative features in the shared feature space can significantly boost the performance of deep domain adaptation methods.

An unsupervised deep domain adaptation approach for robust speech recognition

This paper addresses the robust speech recognition problem as a domain adaptation task. Specifically, we introduce an unsupervised deep domain adaptation (DDA) approach to acoustic modeling in order to eliminate the training–testing mismatch that is common in real-world use of speech recognition. Under a multi-task learning framework, the approach jointly learns two discriminative classifiers using one deep neural network (DNN). As the main task, a label predictor predicts phoneme labels and is used during training and at test time. As the second task, a domain classifier discriminates between the source and the target domains during training. The network is optimized by minimizing the loss of the label classifier and to maximize the loss of the domain classifier at the same time. The proposed approach is easy to implement by modifying a common feed-forward network. Moreover, this unsupervised approach only needs labeled training data from the source domain and some unlabeled raw data of the new domain. Speech recognition experiments on noise/channel distortion and domain shift confirm the effectiveness of the proposed approach. For instance, on the Aurora-4 corpus, compared with the acoustic model trained only using clean data, the DDA approach achieves relative 37.8% word error rate (WER) reduction.