arXiv subjects

He Zhao

Publications and source records attributed to He Zhao.

At least 91 records · Page 5Linked to original sources

arXiv2024

PTaRL: Prototype-based Tabular Representation Learning via Space Calibration

Tabular data have been playing a mostly important role in diverse real-world fields, such as healthcare, engineering, finance, etc. With the recent success of deep learning, many tabular machine learning (ML) methods based on deep networks (e.g., Transformer, ResNet) have achieved competitive performance on tabular benchmarks. However, existing deep tabular ML methods suffer from the representation entanglement and localization, which largely hinders their prediction performance and leads to performance inconsistency on tabular tasks. To overcome these problems, we explore a novel direction of applying prototype learning for tabular ML and propose a prototype-based tabular representation learning framework, PTaRL, for tabular prediction tasks. The core idea of PTaRL is to construct prototype-based projection space (P-Space) and learn the disentangled representation around global data prototypes. Specifically, PTaRL mainly involves two stages: (i) Prototype Generation, that constructs global prototypes as the basis vectors of P-Space for representation, and (ii) Prototype Projection, that projects the data samples into P-Space and keeps the core global data information via Optimal Transport. Then, to further acquire the disentangled representations, we constrain PTaRL with two strategies: (i) to diversify the coordinates towards global prototypes of different representations within P-Space, we bring up a diversification constraint for representation calibration; (ii) to avoid prototype entanglement in P-Space, we introduce a matrix orthogonalization constraint to ensure the independence of global prototypes. Finally, we conduct extensive experiments in PTaRL coupled with state-of-the-art deep tabular ML models on various tabular benchmarks and the results have shown our consistent superiority.

cs.LG↗

arXiv2024

CLIP-DR: Textual Knowledge-Guided Diabetic Retinopathy Grading with Ranking-aware Prompting

Diabetic retinopathy (DR) is a complication of diabetes and usually takes decades to reach sight-threatening levels. Accurate and robust detection of DR severity is critical for the timely management and treatment of diabetes. However, most current DR grading methods suffer from insufficient robustness to data variability (\textit{e.g.} colour fundus images), posing a significant difficulty for accurate and robust grading. In this work, we propose a novel DR grading framework CLIP-DR based on three observations: 1) Recent pre-trained visual language models, such as CLIP, showcase a notable capacity for generalisation across various downstream tasks, serving as effective baseline models. 2) The grading of image-text pairs for DR often adheres to a discernible natural sequence, yet most existing DR grading methods have primarily overlooked this aspect. 3) A long-tailed distribution among DR severity levels complicates the grading process. This work proposes a novel ranking-aware prompting strategy to help the CLIP model exploit the ordinal information. Specifically, we sequentially design learnable prompts between neighbouring text-image pairs in two different ranking directions. Additionally, we introduce a Similarity Matrix Smooth module into the structure of CLIP to balance the class distribution. Finally, we perform extensive comparisons with several state-of-the-art methods on the GDRBench benchmark, demonstrating our CLIP-DR's robustness and superior performance. The implementation code is available \footnote{\url{https://github.com/Qinkaiyu/CLIP-DR}

cs.CV↗

arXiv2024

Towards Generalising Neural Topical Representations

Topic models have evolved from conventional Bayesian probabilistic models to recent Neural Topic Models (NTMs). Although NTMs have shown promising performance when trained and tested on a specific corpus, their generalisation ability across corpora has yet to be studied. In practice, we often expect that an NTM trained on a source corpus can still produce quality topical representation (i.e., latent distribution over topics) for the document from different target corpora to a certain degree. In this work, we aim to improve NTMs further so that their representation power for documents generalises reliably across corpora and tasks. To do so, we propose to enhance NTMs by narrowing the semantic distance between similar documents, with the underlying assumption that documents from different corpora may share similar semantics. Specifically, we obtain a similar document for each training document by text data augmentation. Then, we optimise NTMs further by minimising the semantic distance between each pair, measured by the Topical Optimal Transport (TopicalOT) distance, which computes the optimal transport distance between their topical representations. Our framework can be readily applied to most NTMs as a plug-and-play module. Extensive experiments show that our framework significantly improves the generalisation ability regarding neural topical representation across corpora. Our code and datasets are available at: https://github.com/Xiaohao-Yang/Topic_Model_Generalisation.

cs.CL↗

arXiv2024

Parameter Estimation in DAGs from Incomplete Data via Optimal Transport

Estimating the parameters of a probabilistic directed graphical model from incomplete data is a long-standing challenge. This is because, in the presence of latent variables, both the likelihood function and posterior distribution are intractable without assumptions about structural dependencies or model classes. While existing learning methods are fundamentally based on likelihood maximization, here we offer a new view of the parameter learning problem through the lens of optimal transport. This perspective licenses a general framework that operates on any directed graphs without making unrealistic assumptions on the posterior over the latent variables or resorting to variational approximations. We develop a theoretical framework and support it with extensive empirical evidence demonstrating the versatility and robustness of our approach. Across experiments, we show that not only can our method effectively recover the ground-truth parameters but it also performs comparably or better than competing baselines on downstream applications.

cs.LG↗

arXiv2024

Optimal Transport for Structure Learning Under Missing Data

Causal discovery in the presence of missing data introduces a chicken-and-egg dilemma. While the goal is to recover the true causal structure, robust imputation requires considering the dependencies or, preferably, causal relations among variables. Merely filling in missing values with existing imputation methods and subsequently applying structure learning on the complete data is empirically shown to be sub-optimal. To address this problem, we propose a score-based algorithm for learning causal structures from missing data based on optimal transport. This optimal transport viewpoint diverges from existing score-based approaches that are dominantly based on expectation maximization. We formulate structure learning as a density fitting problem, where the goal is to find the causal model that induces a distribution of minimum Wasserstein distance with the observed data distribution. Our framework is shown to recover the true causal graphs more effectively than competing methods in most simulations and real-data settings. Empirical evidence also shows the superior scalability of our approach, along with the flexibility to incorporate any off-the-shelf causal discovery methods for complete data.

cs.LG↗

arXiv2024

Bayesian Vector AutoRegression with Factorised Granger-Causal Graphs

We study the problem of automatically discovering Granger causal relations from observational multivariate time-series data.Vector autoregressive (VAR) models have been time-tested for this problem, including Bayesian variants and more recent developments using deep neural networks. Most existing VAR methods for Granger causality use sparsity-inducing penalties/priors or post-hoc thresholds to interpret their coefficients as Granger causal graphs. Instead, we propose a new Bayesian VAR model with a hierarchical factorised prior distribution over binary Granger causal graphs, separately from the VAR coefficients. We develop an efficient algorithm to infer the posterior over binary Granger causal graphs. Comprehensive experiments on synthetic, semi-synthetic, and climate data show that our method is more uncertainty aware, has less hyperparameters, and achieves better performance than competing approaches, especially in low-data regimes where there are less observations.

cs.LG↗

arXiv2024

Extracting Clean and Balanced Subset for Noisy Long-tailed Classification

Real-world datasets usually are class-imbalanced and corrupted by label noise. To solve the joint issue of long-tailed distribution and label noise, most previous works usually aim to design a noise detector to distinguish the noisy and clean samples. Despite their effectiveness, they may be limited in handling the joint issue effectively in a unified way. In this work, we develop a novel pseudo labeling method using class prototypes from the perspective of distribution matching, which can be solved with optimal transport (OT). By setting a manually-specific probability measure and using a learned transport plan to pseudo-label the training samples, the proposed method can reduce the side-effects of noisy and long-tailed data simultaneously. Then we introduce a simple yet effective filter criteria by combining the observed labels and pseudo labels to obtain a more balanced and less noisy subset for a robust model training. Extensive experiments demonstrate that our method can extract this class-balanced subset with clean labels, which brings effective performance gains for long-tailed classification with label noise.

cs.LG↗

arXiv2024

Selective, Interpretable, and Motion Consistent Privacy Attribute Obfuscation for Action Recognition

Concerns for the privacy of individuals captured in public imagery have led to privacy-preserving action recognition. Existing approaches often suffer from issues arising through obfuscation being applied globally and a lack of interpretability. Global obfuscation hides privacy sensitive regions, but also contextual regions important for action recognition. Lack of interpretability erodes trust in these new technologies. We highlight the limitations of current paradigms and propose a solution: Human selected privacy templates that yield interpretability by design, an obfuscation scheme that selectively hides attributes and also induces temporal consistency, which is important in action recognition. Our approach is architecture agnostic and directly modifies input imagery, while existing approaches generally require architecture training. Our approach offers more flexibility, as no retraining is required, and outperforms alternatives on three widely used datasets.

cs.CV↗

arXiv2024

Study of the effects caused by space charge in electron cooling

In electron cooling, the space charge (SC) is an important effect, which will affect the e-beam velocity distribution and thus the cooling performance. In this paper, we analyse several important effects that due to the space charge field, such like transverse and longitudianl space charge force, drift velocity caused by SC and magnetic field, and longitudinal momentum deviation due to the SC during acceleration.

physics.acc-ph↗

arXiv2024

Pretext Training Algorithms for Event Sequence Data

Pretext training followed by task-specific fine-tuning has been a successful approach in vision and language domains. This paper proposes a self-supervised pretext training framework tailored to event sequence data. We introduce a novel alignment verification task that is specialized to event sequences, building on good practices in masked reconstruction and contrastive learning. Our pretext tasks unlock foundational representations that are generalizable across different down-stream tasks, including next-event prediction for temporal point process models, event sequence classification, and missing event interpolation. Experiments on popular public benchmarks demonstrate the potential of the proposed method across different tasks and data domains.

cs.LG↗

arXiv2024

Spatial Variations of Dust Opacity and Grain Growth in Dark Clouds: L1689, L1709 and L1712

The far-infrared (FIR) opacity of dust in dark clouds within the Ophiuchus molecular cloud is investigated through multi-wavelength infrared observations from UKIDSS, Spitzer and Herschel. Employing the infrared color excess technique with both near-infrared (NIR) and mid-infrared (MIR) photometric data, a high-resolution extinction map in the $K$ band ($A_K$) is constructed for three dark clouds: L1689, L1709, and L1712. The derived extinction map has a resolution of $1'$ and reaches a depth of $A_K\sim3$ mag. The FIR optical depths $τ_{250}$ at a reference wavelength of $250\,\rm μm$ are obtained by fitting the Herschel PACS and SPIRE continuum data at 100, 160, 250, 350 and 500 $\rm μm$ using a modified blackbody model. The average dust opacity per unit gas mass at $250\rm μm$, $rκ_{250}$ is determined through a pixel-by-pixel correlation of $τ_{250}$ with $A_K$, yielding a value of approximately $0.09\,\rm cm^2\,g^{-1}$, which is about 2-3 times higher than the typical value in the diffuse interstellar medium (ISM). Additionally, an independent analysis across 16 sub-regions within the Ophiuchus cloud indicates spatial variations in dust opacity, with values ranging from 0.07-0.12$\,\rm cm^2\,g^{-1}$. Although the observed trend of increasing dust opacity with higher extinction implies grain growth, our findings indicate that rapid grain growth clearly not yet occurred in the dark clouds studied in this work.

astro-ph.GA↗

arXiv2024

A Class-aware Optimal Transport Approach with Higher-Order Moment Matching for Unsupervised Domain Adaptation

Unsupervised domain adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target domain. In this paper, we introduce a novel approach called class-aware optimal transport (OT), which measures the OT distance between a distribution over the source class-conditional distributions and a mixture of source and target data distribution. Our class-aware OT leverages a cost function that determines the matching extent between a given data example and a source class-conditional distribution. By optimizing this cost function, we find the optimal matching between target examples and source class-conditional distributions, effectively addressing the data and label shifts that occur between the two domains. To handle the class-aware OT efficiently, we propose an amortization solution that employs deep neural networks to formulate the transportation probabilities and the cost function. Additionally, we propose minimizing class-aware Higher-order Moment Matching (HMM) to align the corresponding class regions on the source and target domains. The class-aware HMM component offers an economical computational approach for accurately evaluating the HMM distance between the two distributions. Extensive experiments on benchmark datasets demonstrate that our proposed method significantly outperforms existing state-of-the-art baselines.

cs.LG↗

arXiv2024

Continuous Target Speech Extraction: Enhancing Personalized Diarization and Extraction on Complex Recordings

Target speaker extraction (TSE) aims to extract the target speaker's voice from the input mixture. Previous studies have concentrated on high-overlapping scenarios. However, real-world applications usually meet more complex scenarios like variable speaker overlapping and target speaker absence. In this paper, we introduces a framework to perform continuous TSE (C-TSE), comprising a target speaker voice activation detection (TSVAD) and a TSE model. This framework significantly improves TSE performance on similar speakers and enhances personalization, which is lacking in traditional diarization methods. In detail, unlike conventional TSVAD deployed to refine the diarization results, the proposed Attention-target speaker voice activation detection (A-TSVAD) directly generates timestamps of the target speaker. We also explore some different integration methods of A-TSVAD and TSE by comparing the cascaded and parallel methods. The framework's effectiveness is assessed using a range of metrics, including diarization and enhancement metrics. Our experiments demonstrate that A-TSVAD outperforms conventional methods in reducing diarization errors. Furthermore, the integration of A-TSVAD and TSE in a sequential cascaded manner further enhances extraction accuracy.

cs.SD↗

arXiv2024

HGAttack: Transferable Heterogeneous Graph Adversarial Attack

Heterogeneous Graph Neural Networks (HGNNs) are increasingly recognized for their performance in areas like the web and e-commerce, where resilience against adversarial attacks is crucial. However, existing adversarial attack methods, which are primarily designed for homogeneous graphs, fall short when applied to HGNNs due to their limited ability to address the structural and semantic complexity of HGNNs. This paper introduces HGAttack, the first dedicated gray box evasion attack method for heterogeneous graphs. We design a novel surrogate model to closely resemble the behaviors of the target HGNN and utilize gradient-based methods for perturbation generation. Specifically, the proposed surrogate model effectively leverages heterogeneous information by extracting meta-path induced subgraphs and applying GNNs to learn node embeddings with distinct semantics from each subgraph. This approach improves the transferability of generated attacks on the target HGNN and significantly reduces memory costs. For perturbation generation, we introduce a semantics-aware mechanism that leverages subgraph gradient information to autonomously identify vulnerable edges across a wide range of relations within a constrained perturbation budget. We validate HGAttack's efficacy with comprehensive experiments on three datasets, providing empirical analyses of its generated perturbations. Outperforming baseline methods, HGAttack demonstrated significant efficacy in diminishing the performance of target HGNN models, affirming the effectiveness of our approach in evaluating the robustness of HGNNs against adversarial attacks.

cs.LG↗

arXiv2024

Audio Deepfake Detection with Self-Supervised WavLM and Multi-Fusion Attentive Classifier

With the rapid development of speech synthesis and voice conversion technologies, Audio Deepfake has become a serious threat to the Automatic Speaker Verification (ASV) system. Numerous countermeasures are proposed to detect this type of attack. In this paper, we report our efforts to combine the self-supervised WavLM model and Multi-Fusion Attentive classifier for audio deepfake detection. Our method exploits the WavLM model to extract features that are more conducive to spoofing detection for the first time. Then, we propose a novel Multi-Fusion Attentive (MFA) classifier based on the Attentive Statistics Pooling (ASP) layer. The MFA captures the complementary information of audio features at both time and layer levels. Experiments demonstrate that our methods achieve state-of-the-art results on the ASVspoof 2021 DF set and provide competitive results on the ASVspoof 2019 and 2021 LA set.

eess.AS↗

arXiv2023

Diffuse Interstellar Bands in Gaia DR3 RVS spectra Machine-learning based new measurements

Diffuse interstellar bands (DIBs) are weak and broad interstellar absorption features in astronomical spectra originating from unknown molecules. To measure DIBs in spectra of late-type stars more accurately and more efficiently, we developed a Random Forest model to isolate the DIB features from the stellar components and applied this method to 780 thousand spectra collected by the Gaia Radial Velocity Spectrometer (RVS) that were published in the third data release (DR3). After subtracting the stellar components, we modeled the DIBs $λ$8621 and $λ$8648. After quality control, we selected 7619 reliable measurements. The rest-frame wavelength of DIB $λ$8621 was updated as $λ_0\,{=}\,8623.141\,{\pm}\,0.030$ AA in vacuum, corresponding to 8620.766 AA in air, which was determined by 77 DIB measurements toward the Galactic anti-center. With the peak finding method and a coarse analysis, DIB $λ$8621 was found to correlate better with the neutral hydrogen than the molecular hydrogen (represented by $^{12}$CO $J\,{=}\,(1{-}0)$ emission). We also obtained 179 reliable measurements of DIB $λ$8648 in the RVS spectra of individual stars for the first time, further confirming this very broad DIB feature. A rough estimation of $λ_0$ for DIB $λ$8648 was 8646.31 AA in vacuum, corresponding to 8643.93 AA in air, assuming that the carriers of $λ$8621 and $λ$8648 are co-moving. We confirmed the impact of stellar residuals on the DIB measurements in Gaia DR3, which led to a distortion of the DIB profile and a shift of the center ($\lesssim0.5$ AA), but the EW was consistent with our new measurements. With our measurements and analyses, we propose that the machine-learning-based approach can be widely applied to measure DIBs in numerous spectra from spectroscopic surveys.

astro-ph.GA↗

arXiv2023

Nanoscale strain manipulation of smectic susceptibility in kagome superconductors

Exotic quantum solids can host electronic states that spontaneously break rotational symmetry of the electronic structure, such as electronic nematic phases and unidirectional charge density waves (CDWs). When electrons couple to the lattice, uniaxial strain can be used to anchor and control this electronic directionality. Here we reveal an unusual impact of strain on unidirectional "smectic" CDW orders in kagome superconductors AV3Sb5 using spectroscopic-imaging scanning tunneling microscopy. We discover local decoupling between the smectic electronic director axis and the direction of anisotropic strain. While the two are generally aligned along the same direction in regions of small CDW gap, the two become misaligned in regions where CDW gap is the largest. This in turn suggests nanoscale variations in smectic susceptibility, which we attribute to a combination of local strain and electron correlation strength. Overall, we observe an unusually high decoupling rate between the smectic electronic director of the 3-state Potts order and anisotropic strain, revealing weak smecto-elastic coupling in the CDW phase of kagome superconductors. This is phenomenologically different from the extensively studied nemato-elastic coupling in the Ising nematic phase of Fe-based superconductors, providing a contrasting picture of how strain can control electronic unidirectionality in different families of quantum materials.

cond-mat.supr-con↗

arXiv2023

Dual Conditioned Diffusion Models for Out-Of-Distribution Detection: Application to Fetal Ultrasound Videos

Out-of-distribution (OOD) detection is essential to improve the reliability of machine learning models by detecting samples that do not belong to the training distribution. Detecting OOD samples effectively in certain tasks can pose a challenge because of the substantial heterogeneity within the in-distribution (ID), and the high structural similarity between ID and OOD classes. For instance, when detecting heart views in fetal ultrasound videos there is a high structural similarity between the heart and other anatomies such as the abdomen, and large in-distribution variance as a heart has 5 distinct views and structural variations within each view. To detect OOD samples in this context, the resulting model should generalise to the intra-anatomy variations while rejecting similar OOD samples. In this paper, we introduce dual-conditioned diffusion models (DCDM) where we condition the model on in-distribution class information and latent features of the input image for reconstruction-based OOD detection. This constrains the generative manifold of the model to generate images structurally and semantically similar to those within the in-distribution. The proposed model outperforms reference methods with a 12% improvement in accuracy, 22% higher precision, and an 8% better F1 score.

cs.CV↗