arXiv subjects

Dan Zhang

Publications and source records attributed to Dan Zhang.

At least 163 records · Page 9Linked to original sources

arXiv2023

Anomaly-Aware Semantic Segmentation via Style-Aligned OoD Augmentation

Within the context of autonomous driving, encountering unknown objects becomes inevitable during deployment in the open world. Therefore, it is crucial to equip standard semantic segmentation models with anomaly awareness. Many previous approaches have utilized synthetic out-of-distribution (OoD) data augmentation to tackle this problem. In this work, we advance the OoD synthesis process by reducing the domain gap between the OoD data and driving scenes, effectively mitigating the style difference that might otherwise act as an obvious shortcut during training. Additionally, we propose a simple fine-tuning loss that effectively induces a pre-trained semantic segmentation model to generate a ``none of the given classes" prediction, leveraging per-pixel OoD scores for anomaly segmentation. With minimal fine-tuning effort, our pipeline enables the use of pre-trained models for anomaly segmentation while maintaining the performance on the original task.

cs.CV↗

arXiv2023

A New Family of Perfect Polyphase Sequences with Low Cross-Correlation

Spread spectrum multiple access systems demand minimum possible cross-correlation between the sequences within a set of sequences having good auto-correlation properties. Through a connection between generalised Frank sequences and Florentine arrays, we present a family of perfect sequences with low cross-correlation having a larger family size, compared with previous works. In particular, the family size can be equal to the square root of the period when the period of the perfect sequences is even. In contrast, the number of the perfect sequences of even period with low cross-correlation is equal to one in all previous works.

cs.IT↗

arXiv2023

Search for light dark matter from atmosphere in PandaX-4T

We report a search for light dark matter produced through the cascading decay of $η$ mesons, which are created as a result of inelastic collisions between cosmic rays and Earth's atmosphere. We introduce a new and general framework, publicly accessible, designed to address boosted dark matter specifically, with which a full and dedicated simulation including both elastic and quasi-elastic processes of Earth attenuation effect on the dark matter particles arriving at the detector is performed. In the PandaX-4T commissioning data of 0.63 tonne$\cdot$year exposure, no significant excess over background is observed. The first constraints on the interaction between light dark matter generated in the atmosphere and nucleus through a light scalar mediator are obtained. The lowest excluded cross-section is set at $5.9 \times 10^{-37}{\rm cm^2}$ for dark matter mass of $0.1$ MeV$/c^2$ and mediator mass of 300 MeV$/c^2$. The lowest upper limit of $η$ to dark matter decay branching ratio is $1.6 \times 10^{-7}$.

hep-ex↗

arXiv2023

Contrastive Representation Disentanglement for Clustering

Clustering continues to be a significant and challenging task. Recent studies have demonstrated impressive results by applying clustering to feature representations acquired through self-supervised learning, particularly on small datasets. However, when dealing with datasets containing a large number of clusters, such as ImageNet, current methods struggle to achieve satisfactory clustering performance. In this paper, we introduce a novel method called Contrastive representation Disentanglement for Clustering (CDC) that leverages contrastive learning to directly disentangle the feature representation for clustering. In CDC, we decompose the representation into two distinct components: one component encodes categorical information under an equipartition constraint, and the other component captures instance-specific factors. To train our model, we propose a contrastive loss that effectively utilizes both components of the representation. We conduct a theoretical analysis of the proposed loss and highlight how it assigns different weights to negative samples during the process of disentangling the feature representation. Further analysis of the gradients reveals that larger weights emphasize a stronger focus on hard negative samples. As a result, the proposed loss exhibits strong expressiveness, enabling efficient disentanglement of categorical information. Through experimental evaluation on various benchmark datasets, our method demonstrates either state-of-the-art or highly competitive clustering performance. Notably, on the complete ImageNet dataset, we achieve an accuracy of 53.4%, surpassing existing methods by a substantial margin of +10.2%.

cs.LG↗

arXiv2023

A Search for Light Fermionic Dark Matter Absorption on Electrons in PandaX-4T

We report a search on a sub-MeV fermionic dark matter absorbed by electrons with an outgoing active neutrino using the 0.63 tonne-year exposure collected by PandaX-4T liquid xenon experiment. No significant signals are observed over the expected background. The data are interpreted into limits to the effective couplings between such dark matter and electrons. For axial-vector or vector interactions, our sensitivity is competitive in comparison to existing astrophysical bounds on the decay of such dark matter into photon final states. In particular, we present the first direct detection limits for an axial-vector (vector) interaction which are the strongest in the mass range from 25 to 45 (35 to 50) keV/c$^2$.

hep-ex↗

arXiv2023

Intra- & Extra-Source Exemplar-Based Style Synthesis for Improved Domain Generalization

The generalization with respect to domain shifts, as they frequently appear in applications such as autonomous driving, is one of the remaining big challenges for deep learning models. Therefore, we propose an exemplar-based style synthesis pipeline to improve domain generalization in semantic segmentation. Our method is based on a novel masked noise encoder for StyleGAN2 inversion. The model learns to faithfully reconstruct the image, preserving its semantic layout through noise prediction. Using the proposed masked noise encoder to randomize style and content combinations in the training set, i.e., intra-source style augmentation (ISSA) effectively increases the diversity of training data and reduces spurious correlation. As a result, we achieve up to $12.4\%$ mIoU improvements on driving-scene semantic segmentation under different types of data shifts, i.e., changing geographic locations, adverse weather conditions, and day to night. ISSA is model-agnostic and straightforwardly applicable with CNNs and Transformers. It is also complementary to other domain generalization techniques, e.g., it improves the recent state-of-the-art solution RobustNet by $3\%$ mIoU in Cityscapes to Dark Zürich. In addition, we demonstrate the strong plug-n-play ability of the proposed style synthesis pipeline, which is readily usable for extra-source exemplars e.g., web-crawled images, without any retraining or fine-tuning. Moreover, we study a new use case to indicate neural network's generalization capability by building a stylized proxy validation set. This application has significant practical sense for selecting models to be deployed in the open-world environment. Our code is available at \url{https://github.com/boschresearch/ISSA}.

cs.CV↗

arXiv2023

Transport properties in the Horndeski holographic two-currents model

The transport features of the holographic two-currents model are investigated in the Horndeski gravity framework. This system displays metallic or insulating characteristics depending on whether the Horndeski coupling parameter $γ$ is negative or positive, but is unaffected by other system parameters such as the strength of the momentum dissipation $\hat{k}$, the doping $χ$ and the coupling between two gauge fields $θ$. Secondly, we demonstrate that the thermal conductivities are affected not only by the inherent properties of the black hole, but also by the model parameters. Furthermore, we are particularly interested in the Lorentz ratios' properties. As expected, the Wiedemann-Franz (WF) law is violated, as it is in the majority of holographic systems. Particularly intriguing is the fact that several Lorentz ratio bounds reported in the typical axions model still remain true in our current theories. We would like to highlight out, however, that the lower bound for $\hat{\bar{L}}_A$ is affected by the system parameters $χ$, $θ$ and $γ$, which differs from the case of the typical axions model.

hep-th↗

arXiv2023

Intra-Source Style Augmentation for Improved Domain Generalization

The generalization with respect to domain shifts, as they frequently appear in applications such as autonomous driving, is one of the remaining big challenges for deep learning models. Therefore, we propose an intra-source style augmentation (ISSA) method to improve domain generalization in semantic segmentation. Our method is based on a novel masked noise encoder for StyleGAN2 inversion. The model learns to faithfully reconstruct the image preserving its semantic layout through noise prediction. Random masking of the estimated noise enables the style mixing capability of our model, i.e. it allows to alter the global appearance without affecting the semantic layout of an image. Using the proposed masked noise encoder to randomize style and content combinations in the training set, ISSA effectively increases the diversity of training data and reduces spurious correlation. As a result, we achieve up to $12.4\%$ mIoU improvements on driving-scene semantic segmentation under different types of data shifts, i.e., changing geographic locations, adverse weather conditions, and day to night. ISSA is model-agnostic and straightforwardly applicable with CNNs and Transformers. It is also complementary to other domain generalization techniques, e.g., it improves the recent state-of-the-art solution RobustNet by $3\%$ mIoU in Cityscapes to Dark Zürich.

cs.CV↗

arXiv2023

High Dynamic Range Imaging with Context-aware Transformer

Avoiding the introduction of ghosts when synthesising LDR images as high dynamic range (HDR) images is a challenging task. Convolutional neural networks (CNNs) are effective for HDR ghost removal in general, but are challenging to deal with the LDR images if there are large movements or oversaturation/undersaturation. Existing dual-branch methods combining CNN and Transformer omit part of the information from non-reference images, while the features extracted by the CNN-based branch are bound to the kernel size with small receptive field, which are detrimental to the deblurring and the recovery of oversaturated/undersaturated regions. In this paper, we propose a novel hierarchical dual Transformer method for ghost-free HDR (HDT-HDR) images generation, which extracts global features and local features simultaneously. First, we use a CNN-based head with spatial attention mechanisms to extract features from all the LDR images. Second, the LDR features are delivered to the Hierarchical Dual Transformer (HDT). In each Dual Transformer (DT), the global features are extracted by the window-based Transformer, while the local details are extracted using the channel attention mechanism with deformable CNNs. Finally, the ghost free HDR image is obtained by dimensional mapping on the HDT output. Abundant experiments demonstrate that our HDT-HDR achieves the state-of-the-art performance among existing HDR ghost removal methods.

cs.CV↗

arXiv2023

Identification of Systematic Errors of Image Classifiers on Rare Subgroups

Despite excellent average-case performance of many image classifiers, their performance can substantially deteriorate on semantically coherent subgroups of the data that were under-represented in the training data. These systematic errors can impact both fairness for demographic minority groups as well as robustness and safety under domain shift. A major challenge is to identify such subgroups with subpar performance when the subgroups are not annotated and their occurrence is very rare. We leverage recent advances in text-to-image models and search in the space of textual descriptions of subgroups ("prompts") for subgroups where the target model has low performance on the prompt-conditioned synthesized data. To tackle the exponentially growing number of subgroups, we employ combinatorial testing. We denote this procedure as PromptAttack as it can be interpreted as an adversarial attack in a prompt space. We study subgroup coverage and identifiability with PromptAttack in a controlled setting and find that it identifies systematic errors with high accuracy. Thereupon, we apply PromptAttack to ImageNet classifiers and identify novel systematic errors on rare subgroups.

cs.CV↗

arXiv2023

MM-BSN: Self-Supervised Image Denoising for Real-World with Multi-Mask based on Blind-Spot Network

Recent advances in deep learning have been pushing image denoising techniques to a new level. In self-supervised image denoising, blind-spot network (BSN) is one of the most common methods. However, most of the existing BSN algorithms use a dot-based central mask, which is recognized as inefficient for images with large-scale spatially correlated noise. In this paper, we give the definition of large-noise and propose a multi-mask strategy using multiple convolutional kernels masked in different shapes to further break the noise spatial correlation. Furthermore, we propose a novel self-supervised image denoising method that combines the multi-mask strategy with BSN (MM-BSN). We show that different masks can cause significant performance differences, and the proposed MM-BSN can efficiently fuse the features extracted by multi-masked layers, while recovering the texture structures destroyed by multi-masking and information transmission. Our MM-BSN can be used to address the problem of large-noise denoising, which cannot be efficiently handled by other BSN methods. Extensive experiments on public real-world datasets demonstrate that the proposed MM-BSN achieves state-of-the-art performance among self-supervised and even unpaired image denoising methods for sRGB images denoising, without any labelling effort or prior knowledge. Code can be found in https://github.com/dannie125/MM-BSN.

cs.CV↗

arXiv2023

Self-Supervised Image Denoising for Real-World Images with Context-aware Transformer

In recent years, the development of deep learning has been pushing image denoising to a new level. Among them, self-supervised denoising is increasingly popular because it does not require any prior knowledge. Most of the existing self-supervised methods are based on convolutional neural networks (CNN), which are restricted by the locality of the receptive field and would cause color shifts or textures loss. In this paper, we propose a novel Denoise Transformer for real-world image denoising, which is mainly constructed with Context-aware Denoise Transformer (CADT) units and Secondary Noise Extractor (SNE) block. CADT is designed as a dual-branch structure, where the global branch uses a window-based Transformer encoder to extract the global information, while the local branch focuses on the extraction of local features with small receptive field. By incorporating CADT as basic components, we build a hierarchical network to directly learn the noise distribution information through residual learning and obtain the first stage denoised output. Then, we design SNE in low computation for secondary global noise extraction. Finally the blind spots are collected from the Denoise Transformer output and reconstructed, forming the final denoised image. Extensive experiments on the real-world SIDD benchmark achieve 50.62/0.990 for PSNR/SSIM, which is competitive with the current state-of-the-art method and only 0.17/0.001 lower. Visual comparisons on public sRGB, Raw-RGB and greyscale datasets prove that our proposed Denoise Transformer has a competitive performance, especially on blurred textures and low-light images, without using additional knowledge, e.g., noise level or noise type, regarding the underlying unknown noise.

cs.CV↗

arXiv2023

Towards Transparent, Reusable, and Customizable Data Science in Computational Notebooks

Data science workflows are human-centered processes involving on-demand programming and analysis. While programmable and interactive interfaces such as widgets embedded within computational notebooks are suitable for these workflows, they lack robust state management capabilities and do not support user-defined customization of the interactive components. The absence of such capabilities hinders workflow reusability and transparency while limiting the scope of exploration of the end-users. In response, we developed MAGNETON, a framework for authoring interactive widgets within computational notebooks that enables transparent, reusable, and customizable data science workflows. The framework enhances existing widgets to support fine-grained interaction history management, reusable states, and user-defined customizations. We conducted three case studies in a real-world knowledge graph construction and serving platform to evaluate the effectiveness of these widgets. Based on the observations, we discuss future implications of employing MAGNETON widgets for general-purpose data science workflows.

cs.HC↗

arXiv2023

Learning From Yourself: A Self-Distillation Method for Fake Speech Detection

In this paper, we propose a novel self-distillation method for fake speech detection (FSD), which can significantly improve the performance of FSD without increasing the model complexity. For FSD, some fine-grained information is very important, such as spectrogram defects, mute segments, and so on, which are often perceived by shallow networks. However, shallow networks have much noise, which can not capture this very well. To address this problem, we propose using the deepest network instruct shallow network for enhancing shallow networks. Specifically, the networks of FSD are divided into several segments, the deepest network being used as the teacher model, and all shallow networks become multiple student models by adding classifiers. Meanwhile, the distillation path between the deepest network feature and shallow network features is used to reduce the feature difference. A series of experimental results on the ASVspoof 2019 LA and PA datasets show the effectiveness of the proposed method, with significant improvements compared to the baseline.

cs.SD↗

arXiv2023

Quasinormal modes and Hawking radiation of a charged Weyl black hole

We investigate the scalar field system over a charged Weyl black hole, depicted by a parameter $λ$. It is found that the imaginary part of the quasinormal mode spectra is always negative and the perturbation does not increase with the time, indicating that the system is stable under scalar field perturbation. Furthermore, the quasinormal mode spectra and Hawking radiation exhibit a qualitatively similar characteristic in that they both rise with rising $λ$ and approach a constant when $λ$ is large enough. Especially, we would like to emphasize that an exponential decay obviously emerges in the phase of the ringing tail as $λ$ increases. It indicates that the characteristic parameter $λ$ has an obvious imprint on the ringing tail, which is expected to be detected by future observations.

gr-qc↗

arXiv2023

A Data Driven Method for Multi-step Prediction of Ship Roll Motion in High Sea States

Ship roll motion in high sea states has large amplitudes and nonlinear dynamics, and its prediction is significant for operability, safety, and survivability. This paper presents a novel data-driven methodology to provide a multi-step prediction of ship roll motions in high sea states. A hybrid neural network is proposed that combines long short-term memory (LSTM) and convolutional neural network (CNN) in parallel. The motivation is to extract the nonlinear dynamic characteristics and the hydrodynamic memory information through the advantage of CNN and LSTM, respectively. For the feature selection, the time histories of motion states and wave heights are selected to involve sufficient information. Taken a scaled KCS as the study object, the ship motions in sea state 7 irregular long-crested waves are simulated and used for the validation. The results show that at least one period of roll motion can be accurately predicted. Compared with the single LSTM and CNN methods, the proposed method has better performance in predicting the amplitude of roll angles. Besides, the comparison results also demonstrate that selecting motion states and wave heights as feature space improves the prediction accuracy, verifying the effectiveness of the proposed method.

cs.LG↗

arXiv2023

GOOD: Exploring Geometric Cues for Detecting Objects in an Open World

We address the task of open-world class-agnostic object detection, i.e., detecting every object in an image by learning from a limited number of base object classes. State-of-the-art RGB-based models suffer from overfitting the training classes and often fail at detecting novel-looking objects. This is because RGB-based models primarily rely on appearance similarity to detect novel objects and are also prone to overfitting short-cut cues such as textures and discriminative parts. To address these shortcomings of RGB-based object detectors, we propose incorporating geometric cues such as depth and normals, predicted by general-purpose monocular estimators. Specifically, we use the geometric cues to train an object proposal network for pseudo-labeling unannotated novel objects in the training set. Our resulting Geometry-guided Open-world Object Detector (GOOD) significantly improves detection recall for novel object categories and already performs well with only a few training classes. Using a single "person" class for training on the COCO dataset, GOOD surpasses SOTA methods by 5.0% AR@100, a relative improvement of 24%.

cs.CV↗

arXiv2023

Semantics-aware Dataset Discovery from Data Lakes with Contextualized Column-based Representation Learning

Dataset discovery from data lakes is essential in many real application scenarios. In this paper, we propose Starmie, an end-to-end framework for dataset discovery from data lakes (with table union search as the main use case). Our proposed framework features a contrastive learning method to train column encoders from pre-trained language models in a fully unsupervised manner. The column encoder of Starmie captures the rich contextual semantic information within tables by leveraging a contrastive multi-column pre-training strategy. We utilize the cosine similarity between column embedding vectors as the column unionability score and propose a filter-and-verification framework that allows exploring a variety of design choices to compute the unionability score between two tables accordingly. Empirical evaluation results on real table benchmark datasets show that Starmie outperforms the best-known solutions in the effectiveness of table union search by 6.8 in MAP and recall. Moreover, Starmie is the first to employ the HNSW (Hierarchical Navigable Small World) index for accelerate query processing of table union search which provides a 3,000X performance gain over the linear scan baseline and a 400X performance gain over an LSH index (the state-of-the-art solution for data lake indexing).

cs.DB↗