Domain Alignment Research Articles

Enhanced Electro‐Optic Coefficients in Single‐Crystalline BaTiO3 Thin Films Enabled by Domain Alignment

AbstractIn silicon photonics, BaTiO3 (BTO) emerges as a promising electro‐optic (EO) integrated material due to its excellent EO modulation capabilities. Nevertheless, the EO response of multi‐domain BTO thin films tends to be lower than that of bulk materials. BTO thin films are fabricated with strong EO responses utilizing the pulsed laser deposition technique. Considering the importance of the SrTiO3 (STO) buffer layer, a series of epitaxial BTO thin films are deposited on STO substrates with engineered domain alignment. Given the contribution of domain variants in BTO thin films to their EO response, specific electrode patterns are designed to enhance their performance. An effective Pockels coefficient of 286 pm V−1 is determined for the BTO thin film via the transmission geometry method, which surpasses recent reports. Through domain engineering in the BTO thin film, the remarkable enhancement in the EO response arising from in‐plane polarization (a‐axis) characteristics is confirmed. Simultaneously, utilizing the Second Harmonic Generation (SHG) optical probe, deep exploration into the domain dynamics within BTO thin films is conducted. These findings provide insights into the contribution of domain variants to the EO response and shed light on further developing silicon‐based heterogeneously integrated devices.

DDAGCN: an unsupervised cross-domain identification method for tie rod bolt loosening in a rod-fastening rotor system under different working conditions

The cross-domain diagnosis of tie rod bolt loosening is essential for guaranteeing the healthy operation of rod-fastening rotor (RFR) systems. The unsupervised domain adaptation (UDA) method effectively alleviates the impact of domain discrepancy and has been applied for cross-domain diagnosis. Traditional UDA methods mainly focus on the marginal and conditional distributions with fixed weights to adapt the domain distribution discrepancy. However, the fixed distribution combination cannot satisfy the requirement of feature domain alignment under different working conditions, and the relative importance of the two distributions cannot be evaluated quantitatively. This paper proposes an improved dynamic distribution adaptive graph convolutional network (DDAGCN) for the cross-domain diagnosis of tie rod bolt loosening under different working conditions. This method can quantitatively evaluate the relative significance of each distribution in representing the distribution discrepancy. First, it combines the convolutional neural network and the graph convolutional network to extract the features in the graph structure by using the connection relationship between nodes, and realizes the full extraction of neighbourhood information of nodes. Then, the dynamic distribution adaptive alignment strategy is introduced to construct the dynamic linear combination of marginal and conditional distributions, so as to measure the distribution discrepancy between domains. Meanwhile, the domain adversarial module is combined to further reduce the domain gap and finally realize feature alignment. The extracted domain invariant features can effectively enhance the generalization ability and fault identification ability of the model. The case of the public bearing dataset verifies that the effectiveness and generalization ability of the proposed method for cross-domain fault diagnosis under different working conditions is superior to other compared methods. In addition, the identification ability of the proposed method for the degree of tie rod bolt loosening is verified by the self-made bolt loosening dataset of the RFR system.

Domain-adaptive person re-identification via domain alignment and mutual pseudo-label refinement

Domain-adaptive person re-identification via domain alignment and mutual pseudo-label refinement

Atom-by-atom imaging of moiré transformations in 2D transition metal dichalcogenides.

Understanding the atomic-scale mechanisms that govern the structure of interfaces is critical across materials systems but particularly so for two-dimensional (2D) moiré materials. Here, we image, atom-by-atom, the thermally induced structural evolution of twisted bilayer transition metal dichalcogenides using in situ transmission electron microscopy. We observe low-temperature, local conversion of moiré superlattice into nanoscale aligned domains. Unexpectedly, this process occurs by nucleating a new grain within one monolayer, whose crystal orientation is templated by the other. The aligned domains grow through collective rotation of moiré supercells and hopping of 5|7 defect pairs at moiré boundaries. This provides mechanistic insight into the atomic-scale interactions controlling moiré structures and illustrates the potential to pattern interfacial structure and properties of 2D materials at the nanoscale.

A gradient aligned domain adversarial network for unsupervised intelligent fault diagnosis of gearboxes

Unsupervised domain adaptation alleviates the dependencies of conventional fault diagnosis methods on sufficient labeled data and strict data distributions. Nonetheless, the current domain adaptation methods only concentrate on the data distributions and ignore the feature gradient distributions, leading to some samples being misclassified due to large gradient discrepancies, thus affecting diagnosis performance. In this paper, a gradient aligned domain adversarial network (GADAN) is proposed. First, the discrepancies of the marginal and conditional distribution between the source and target domain are reduced by minimizing the joint maximum mean discrepancy. Then, a pseudo-labeling approach based on a clustering self-supervised strategy is utilized to attain high-quality pseudo-labels of target domains, and most importantly in the dimension of the data gradient, the feature gradient distributions are aligned by adversarial learning to further reduce the domain shift, even if the distributions of the two domains are close enough. Finally, experiments and engineering applications demonstrate the effectiveness and superiority of GADAN for transfer diagnosis between various working conditions or different machines.

Patch-Based Auxiliary Node Classification for Domain Adaptive Object Detection

Domain adaptive object detection (DAOD) aims to leverage labeled source domain data to train object detection models that can generalize well to unlabeled target domains. Recently, many researchers have considered implementing fine-grained pixel-level domain adaptation using graph representations. Existing methods construct semantically complete graphs and align them across domains via graph matching. This work introduced an auxiliary node classification task before domain alignment through graph matching, which utilizes the inherent information of graph nodes to classify them, in order to avoid suboptimal graph matching results caused by node class confusion. However, previous methods neglected the contextual information of graph nodes, leading to biased node classification and suboptimal graph matching. To solve this issue, we propose a novel patch-based auxiliary node classification method for DAOD. Unlike existing methods that use only the inherent information of nodes for node classification, our method exploits the local region information of nodes and employs multi-layer convolutional neural networks to learn the local region feature representation of nodes, enriching the node context information. Thus, accurate and robust node classification results are produced and the risk of class confusion is reduced. Moreover, we propose a progressive strategy to fuse the inherent features and the learned local region features of nodes, which ensures that the network can stably and reliably utilize local region features for accurate node classification. In this paper, we conduct abundant experiments on various DAOD scenarios and demonstrate that our proposed model outperforms existing works.

Dual Mode Optoelectronic Devices Based on 2D Single Crystalline VO2 Films with Controlled Metallic Domain Regime Induced Variable Hysteresis

AbstractThe multimode optoelectronic operation in a single component device is an attractive design strategy for future optoelectronics, which demands unique and complex multifunction. Here, a novel dual‐mode optoelectronic device with both memory and switching functionalities by utilizing alternating metallic stripe domain regime‐induced controllable hysteresis of two dimensional (2D)‐crystalline VO2 films is reported. It is found that the formation and disappearance of tensile strain‐induced metallic stripe domain arrays are susceptible to temperature‐induced thermal energy, allowing controllable hysteresis in electric field‐induced metal‐insulator transition (E‐MIT) by tuning the total region of aligned metallic stripe domains within the VO2 films through current flow‐induced thermal energy. Based on the tunable hysteresis in E‐MIT, the 2D‐crystalline VO2 film‐based device exhibits successful multimode optoelectronic memory and switching operations. The device with a large hysteresis width exhibits high‐performance optoelectronic memory behavior with a high on/off ratio of up to ≈55% and a long retention time over 5000 s response to a light pulse optically triggering MIT. On the other hand, the 2D‐crystalline VO2 film device with a narrow hysteresis width exhibits high performance of photodetection with a responsivity of 316 mA W−1 and response times of ≈1.2 and ≈2 µs at rise and fall, respectively.

DANNMCTG: Domain-Adversarial Training of Neural Network for multicenter antenatal cardiotocography signal classification

Intelligent classification of cardiotocography (CTG) based on machine learning (ML), a useful tool to improve the accuracy of fetal abnormality detection, can assist obstetricians with clinical decisions. With the advancement of information technologies and medical devices, there are development opportunities for multicenter clinical research and obtaining more digital CTG signals. However, most of the existing clinical multicenter CTG datasets are partially annotated and have discrepancies which do not satisfy the ML condition of independent identical distribution. Therefore, this paper focuses on an unsupervised domain adaptation (UDA) algorithm to realize cross-domain intelligent classification of multimodal CTG signals. We propose a method dubbed domain adversarial training of neural network for multicenter CTG (DANNMCTG), which mainly consists of a label classifier, a feature extractor and a domain discriminator. To match different distribution of fetal heart rate (FHR), uterine contraction (UC) and fetal movement (FetMov) signals, we condition the domain alignment on label predictions by defining the multi-linear map. For analysis, two datasets from the hospital central station and home monitoring devices were considered as the source and target domains. The results showed that the accuracy value, F1 value and area under the curve (AUC) value of the DANNMCTG were 71.25%, 76.08% and 0.7705, respectively. This method significantly improved the performance of the deep learning models without exploiting any information in the target domain, and outperformed the state-of-the-art UDA algorithms for CTG classification. In summary, the DANNMCTG can effectively mitigate the influence of domain shift for multicenter intelligent prenatal fetal monitoring.

DiDA: Disambiguated Domain Alignment for Cross-Domain Retrieval with Partial Labels

Driven by generative AI and the Internet, there is an increasing availability of a wide variety of images, leading to the significant and popular task of cross-domain image retrieval. To reduce annotation costs and increase performance, this paper focuses on an untouched but challenging problem, i.e., cross-domain image retrieval with partial labels (PCIR). Specifically, PCIR faces great challenges due to the ambiguous supervision signal and the domain gap. To address these challenges, we propose a novel method called disambiguated domain alignment (DiDA) for cross-domain retrieval with partial labels. In detail, DiDA elaborates a novel prototype-score unitization learning mechanism (PSUL) to extract common discriminative representations by simultaneously disambiguating the partial labels and narrowing the domain gap. Additionally, DiDA proposes a prototype-based domain alignment mechanism (PBDA) to further bridge the inherent cross-domain discrepancy. Attributed to PSUL and PBDA, our DiDA effectively excavates domain-invariant discrimination for cross-domain image retrieval. We demonstrate the effectiveness of DiDA through comprehensive experiments on three benchmarks, comparing it to existing state-of-the-art methods. Code available: https://github.com/lhrrrrrr/DiDA.

XKD: Cross-Modal Knowledge Distillation with Domain Alignment for Video Representation Learning

We present XKD, a novel self-supervised framework to learn meaningful representations from unlabelled videos. XKD is trained with two pseudo objectives. First, masked data reconstruction is performed to learn modality-specific representations from audio and visual streams. Next, self-supervised cross-modal knowledge distillation is performed between the two modalities through a teacher-student setup to learn complementary information. We introduce a novel domain alignment strategy to tackle domain discrepancy between audio and visual modalities enabling effective cross-modal knowledge distillation. Additionally, to develop a general-purpose network capable of handling both audio and visual streams, modality-agnostic variants of XKD are introduced, which use the same pretrained backbone for different audio and visual tasks. Our proposed cross-modal knowledge distillation improves video action classification by 8% to 14% on UCF101, HMDB51, and Kinetics400. Additionally, XKD improves multimodal action classification by 5.5% on Kinetics-Sound. XKD shows state-of-the-art performance in sound classification on ESC50, achieving top-1 accuracy of 96.5%.

Triple Feature Disentanglement for One-Stage Adaptive Object Detection

In recent advancements concerning Domain Adaptive Object Detection (DAOD), unsupervised domain adaptation techniques have proven instrumental. These methods enable enhanced detection capabilities within unlabeled target domains by mitigating distribution differences between source and target domains. A subset of DAOD methods employs disentangled learning to segregate Domain-Specific Representations (DSR) and Domain-Invariant Representations (DIR), with ultimate predictions relying on the latter. Current practices in disentanglement, however, often lead to DIR containing residual domain-specific information. To address this, we introduce the Multi-level Disentanglement Module (MDM) that progressively disentangles DIR, enhancing comprehensive disentanglement. Additionally, our proposed Cyclic Disentanglement Module (CDM) facilitates DSR separation. To refine the process further, we employ the Categorical Features Disentanglement Module (CFDM) to isolate DIR and DSR, coupled with category alignment across scales for improved source-target domain alignment. Given its practical suitability, our model is constructed upon the foundational framework of the Single Shot MultiBox Detector (SSD), which is a one-stage object detection approach. Experimental validation highlights the effectiveness of our method, demonstrating its state-of-the-art performance across three benchmark datasets.

Contrastive domain-adaptive graph selective self-training network for cross-network edge classification

The performance of graph Neural Network (GNNs) can degrade significantly when trained on the graphs with noisy edges connecting nodes from different classes. To mitigate the negative effect of noisy edges, previous studies have largely focused on predicting the label agreement between node pairs within a single network. So far, predicting noisy edges across different networks remains largely underexplored. To bridge this gap, our work studies a novel problem of cross-network homophilous and heterophilous edge classification (CNHHEC), aiming to predict the label agreement of edges in an unlabeled target network by transferring the knowledge from a labeled source network. A novel Contrastive Domain-adaptive Graph Self-training Network (CDGSN) is proposed. Firstly, CDGSN learns node and edge embeddings end-to-end by a GNN encoder with adaptive edge weights during neighborhood aggregation. Secondly, to facilitate knowledge transfer across networks, CDGSN employs an adversarial domain adaptation module to align edge embeddings across networks, and also designs a novel contrastive domain adaptation module to conduct class-aware cross-network alignment of node embeddings. As a result, the intra-class domain divergence can be mitigated while the inter-class domain discrepancy can be enlarged to yield network-invariant and label-discriminative node and edge embeddings. Moreover, CDGSN designs a selective positive and negative pseudo-labeling strategy to assign positive (negative) pseudo-labels to the target nodes with extremely high (low) prediction confidence of belonging to each specific class. Such pseudo-labeled target nodes would be employed to iteratively re-train the model in a self-training manner, so as to obtain more reliable target pseudo-labels progressively to guide class-aware domain alignment. Extensive experiments on benchmark datasets demonstrate the effectiveness of the proposed CDGSN on the CNHHEC problem. The performance of CDGSN is robust against various types of edge embeddings. When adopting three operators to construct edge embeddings, the proposed CDGSN can improve the state-of-the-art method for CNHHEC by an average of 0.5 %, 2.8 %, and 10.8 % in terms of AUC, and 1.3 %, 3.0 %, and 6.3 % in terms of AP, respectively.

Open-Set Graph Domain Adaptation via Separate Domain Alignment

Domain adaptation has become an attractive learning paradigm, as it can leverage source domains with rich labels to deal with classification tasks in an unlabeled target domain. A few recent studies develop domain adaptation approaches for graph-structured data. In the case of node classification task, current domain adaptation methods only focus on the closed-set setting, where source and target domains share the same label space. A more practical assumption is that the target domain may contain new classes that are not included in the source domain. Therefore, in this paper, we introduce a novel and challenging problem for graphs, i.e., open-set domain adaptive node classification, and propose a new approach to solve it. Specifically, we develop an algorithm for efficient knowledge transfer from a labeled source graph to an unlabeled target graph under a separate domain alignment (SDA) strategy, in order to learn discriminative feature representations for the target graph. Our goal is to not only correctly classify target nodes into the known classes, but also classify unseen types of nodes into an unknown class. Experimental results on real-world datasets show that our method outperforms existing methods on graph domain adaptation.

Effective Comparative Prototype Hashing for Unsupervised Domain Adaptation

Unsupervised domain adaptive hashing is a highly promising research direction within the field of retrieval. It aims to transfer valuable insights from the source domain to the target domain while maintaining high storage and retrieval efficiency. Despite its potential, this field remains relatively unexplored. Previous methods usually lead to unsatisfactory retrieval performance, as they frequently directly apply slightly modified domain adaptation algorithms to hash learning framework, or pursue domain alignment within the Hamming space characterized by limited semantic information. In this paper, we propose a simple yet effective approach named Comparative Prototype Hashing (CPH) for unsupervised domain adaptive image retrieval. We establish a domain-shared unit hypersphere space through prototype contrastive learning and then obtain the Hamming hypersphere space via mapping from the shared hypersphere. This strategy achieves a cohesive synergy between learning uniformly distributed and category conflict-averse feature representations, eliminating domain discrepancies, and facilitating hash code learning. Moreover, by leveraging dual-domain information to supervise the entire hashing model training process, we can generate hash codes that retain inter-sample similarity relationships within both domains. Experimental results validate that our CPH significantly outperforms the state-of-the-art counterparts across multiple cross-domain and single-domain retrieval tasks. Notably, on Office-Home and Office-31 datasets, CPH achieves an average performance improvement of 19.29% and 13.85% on cross-domain retrieval tasks compared to the second-best results, respectively. The source codes of our method are available at: https://github.com/christinecui/CPH.

An Efficient Knowledge Transfer Strategy for Spiking Neural Networks from Static to Event Domain

Spiking neural networks (SNNs) are rich in spatio-temporal dynamics and are suitable for processing event-based neuromorphic data. However, event-based datasets are usually less annotated than static datasets. This small data scale makes SNNs prone to overfitting and limits their performance. In order to improve the generalization ability of SNNs on event-based datasets, we use static images to assist SNN training on event data. In this paper, we first discuss the domain mismatch problem encountered when directly transferring networks trained on static datasets to event data. We argue that the inconsistency of feature distributions becomes a major factor hindering the effective transfer of knowledge from static images to event data. To address this problem, we propose solutions in terms of two aspects: feature distribution and training strategy. Firstly, we propose a knowledge transfer loss, which consists of domain alignment loss and spatio-temporal regularization. The domain alignment loss learns domain-invariant spatial features by reducing the marginal distribution distance between the static image and the event data. Spatio-temporal regularization provides dynamically learnable coefficients for domain alignment loss by using the output features of the event data at each time step as a regularization term. In addition, we propose a sliding training strategy, which gradually replaces static image inputs probabilistically with event data, resulting in a smoother and more stable training for the network. We validate our method on neuromorphic datasets, including N-Caltech101, CEP-DVS, and N-Omniglot. The experimental results show that our proposed method achieves better performance on all datasets compared to the current state-of-the-art methods. Code is available at https://github.com/Brain-Cog-Lab/Transfer-for-DVS.

Improved Adversarial Transfer Network for Bearing Fault Diagnosis under Variable Working Conditions

Bearings are one of the critical components of rotating machinery, and their failure can cause catastrophic consequences. In this regard, previous studies have proposed a variety of intelligent diagnosis methods. Most existing bearing fault diagnosis methods implicitly assume that the training and test sets are from the same distribution. However, in real scenarios, bearings have been working in complex and changeable working environments for a long time. The data during their working processes and the data used for model training cannot meet this condition. This paper proposes an improved adversarial transfer network for fault diagnosis under variable working conditions. Specifically, this paper combines an adversarial transfer network with a short-time Fourier transform to obtain satisfactory results with the lighter network. Then, this paper employs a channel attention module to enhance feature fusion. Moreover, this paper designs a novel domain discrepancy hybrid metric loss to improve model transfer learning performance. Finally, this paper verifies the method’s effectiveness on three datasets, including dual-rotor, a Case Western Reserve University dataset and the Ottawa dataset. The proposed method achieves average accuracy, surpassing other methods, and shows better domain alignment capabilities.

SOmicsFusion: Multimodal coregistration and fusion between spatial metabolomics and biomedical imaging

We present SOmicsFusion, a software toolbox for ’fusing’ spatial omics with classical biomedical imaging modalities, capitalizing on their inherent correspondences and complementarity when characterizing the same subject. By augmenting radiological and histological images with spatially resolved molecular profiling, this fusion offers a panoramic characterization of the biochemical perturbations underlying pathological conditions, thereby advancing our understanding of diseases like brain disorders and cancers. The cornerstone of SOmicsFusion is a coregistration tool that leverages an innovative two-stage machine learning pipeline to tackle the longstanding challenge of spatially aligning data from fundamentally different modalities, priming them for subsequent fusion analysis that often requires precise pixel-wise correspondence between the datasets. Specifically, the pipeline utilizes an original dimension reduction algorithm for representational domain alignment, followed by a Deep Learning-based method for spatial domain alignment. SOmicsFusion is demonstrated using mass spectrometry imaging (MSI)-mediated spatial metabolomics and four other modalities: magnetic resonance imaging (MRI), microscopy, brain atlas, and spatial transcriptomics. By reducing coregistration errors by 38–69% compared to existing pipelines, SOmicsFusion enhances the precision of associating molecule distribution with anatomy and pathology features, ultimately leading to more statistically robust findings. Furthermore, SOmicsFusion incorporates various downstream analysis tools, including overlay visualization, spatial correlation/co-expression analysis, pansharpening, and automated anatomy annotation. These tools facilitate the extraction of biological insights that would be unattainable through individual modalities alone. For instance, the coregistration and correlation between MSI and in vivo MRI datasets unveil that the spatial heterogeneity in metabolites stems from the temporal heterogeneity in the development of cerebral ischemia-reperfusion injury.

Deep Joint Semantic Adaptation Network for Multi-source Unsupervised Domain Adaptation

Multi-source Unsupervised Domain Adaptation (MUDA) transfers knowledge learned from multiple labeled source domains to an unlabeled target domain by minimizing the domain shift between multiple source domains and the target domain. Recent studies on MUDA have focused on aligning the distribution of each pair of source and target domains in separate feature spaces to reduce their domain shift. However, these approaches suffer from two main shortcomings. First, they usually focus on the global domain shift which lacks consideration of the joint distribution of category-corresponded subdomains. Second, out-of-distribution samples far from the sample center are hard to align by the global domain alignment. Therefore, we propose a novel Deep Joint Semantic Adaptive Network (DJSAN) for MUDA. Specifically, a new maximum mean discrepancy-based metric, Joint Semantic Maximum Mean Discrepancy (JSMMD), is proposed, which can uniformly optimize the cross-domain joint distribution of category-corresponded subdomains on multiple task-specific layers. Moreover, to deal with the out-of-distribution hard samples, we propose an across-domain data augmentation method called Source-Target Domain Mixing (STDMix) to enhance the robustness of the model, which synthesizes the source domain and target domain into a new domain at a fixed ratio and utilizes information entropy to provide reliable pseudo-labels for samples in the target domain. Experimental results on three public datasets, i.e., Office-31, Digits-five, and Office-Home, show that our proposed method achieves improvements of 0.3%, 1.8%, and 2.7% in average accuracy, respectively.

The effect of intrinsic magnetic order on electrochemical water splitting

To reach a long term viable green hydrogen economy, rational design of active oxygen evolution reaction (OER) catalysts is critical. An important hurdle in this reaction originates from the fact that the reactants are singlet molecules, whereas the oxygen molecule has a triplet ground state with parallel spin alignment, implying that magnetic order in the catalyst is essential. Accordingly, multiple experimentalists reported a positive effect of external magnetic fields on OER activity of ferromagnetic catalysts. However, it remains a challenge to investigate the influence of the intrinsic magnetic order on catalytic activity. Here, we tuned the intrinsic magnetic order of epitaxial La0.67Sr0.33MnO3 thin film model catalysts from ferro- to paramagnetic by changing the temperature in situ during water electrolysis. Using this strategy, we show that ferromagnetic ordering below the Curie temperature enhances OER activity. Moreover, we show a slight current density enhancement upon application of an external magnetic field and find that the dependence of magnetic field direction correlates with the magnetic anisotropy in the catalyst film. Our work, thus, suggests that both the intrinsic magnetic order in La0.67Sr0.33MnO3 films and magnetic domain alignment increase their catalytic activity. We observe no long-range magnetic order at the catalytic surface, implying that the OER enhancement is connected to the magnetic order of the bulk catalyst. Combining the effects found with existing literature, we propose a unifying picture for the spin-polarized enhancement in magnetic oxide catalysts.

Multiple Adverse Weather Conditions Adaptation for Object Detection via Causal Intervention.

Most state-of-the-art object detection methods have achieved impressive perfomrace on several public benchmarks, which are trained with high definition images. However, existing detectors are often sensitive to the visual variations and out-of-distribution data due to the domain gap caused by various confounders, e.g. the adverse weathre conditions. To bridge the gap, previous methods have been mainly exploring domain alignment, which requires to collect an amount of domain-specific training samples. In this paper, we introduce a novel domain adaptation model to discover a weather condition invariant feature representation. Specifically, we first employ a memory network to develop a confounder dictionary, which stores prototypes of object features under various scenarios. To guarantee the representativeness of each prototype in the dictionary, a dynamic item extraction strategy is used to update the memory dictionary. After that, we introduce a causal intervention reasoning module to explore the invariant representation of a specific object under different weather conditions. Finally, a categorical consistency regularization is used to constrain the similarities between categories in order to automatically search for the aligned instances among distinct domains. Experiments are conducted on several public benchmarks (RTTS, Foggy-Cityscapes, RID, and BDD 100K) with state-of-the-art performance achieved under multiple weather conditions.