arXiv subjects

Ke Chen

Publications and source records attributed to Ke Chen.

At least 217 records · Page 12Linked to original sources

arXiv2023

Universal Source Separation with Weakly Labelled Data

Universal source separation (USS) is a fundamental research task for computational auditory scene analysis, which aims to separate mono recordings into individual source tracks. There are three potential challenges awaiting the solution to the audio source separation task. First, previous audio source separation systems mainly focus on separating one or a limited number of specific sources. There is a lack of research on building a unified system that can separate arbitrary sources via a single model. Second, most previous systems require clean source data to train a separator, while clean source data are scarce. Third, there is a lack of USS system that can automatically detect and separate active sound classes in a hierarchical level. To use large-scale weakly labeled/unlabeled audio data for audio source separation, we propose a universal audio source separation framework containing: 1) an audio tagging model trained on weakly labeled data as a query net; and 2) a conditional source separation model that takes query net outputs as conditions to separate arbitrary sound sources. We investigate various query nets, source separation models, and training strategies and propose a hierarchical USS strategy to automatically detect and separate sound classes from the AudioSet ontology. By solely leveraging the weakly labelled AudioSet, our USS system is successful in separating a wide variety of sound classes, including sound event separation, music source separation, and speech enhancement. The USS system achieves an average signal-to-distortion ratio improvement (SDRi) of 5.57 dB over 527 sound classes of AudioSet; 10.57 dB on the DCASE 2018 Task 2 dataset; 8.12 dB on the MUSDB18 dataset; an SDRi of 7.28 dB on the Slakh2100 dataset; and an SSNR of 9.00 dB on the voicebank-demand dataset. We release the source code at https://github.com/bytedance/uss

cs.SD↗

arXiv2023

Deep Partial Multi-Label Learning with Graph Disambiguation

In partial multi-label learning (PML), each data example is equipped with a candidate label set, which consists of multiple ground-truth labels and other false-positive labels. Recently, graph-based methods, which demonstrate a good ability to estimate accurate confidence scores from candidate labels, have been prevalent to deal with PML problems. However, we observe that existing graph-based PML methods typically adopt linear multi-label classifiers and thus fail to achieve superior performance. In this work, we attempt to remove several obstacles for extending them to deep models and propose a novel deep Partial multi-Label model with grAph-disambIguatioN (PLAIN). Specifically, we introduce the instance-level and label-level similarities to recover label confidences as well as exploit label dependencies. At each training epoch, labels are propagated on the instance and label graphs to produce relatively accurate pseudo-labels; then, we train the deep model to fit the numerical labels. Moreover, we provide a careful analysis of the risk functions to guarantee the robustness of the proposed model. Extensive experiments on various synthetic datasets and three real-world PML datasets demonstrate that PLAIN achieves significantly superior results to state-of-the-art methods.

cs.LG↗

arXiv2023

SwinFSR: Stereo Image Super-Resolution using SwinIR and Frequency Domain Knowledge

Stereo Image Super-Resolution (stereoSR) has attracted significant attention in recent years due to the extensive deployment of dual cameras in mobile phones, autonomous vehicles and robots. In this work, we propose a new StereoSR method, named SwinFSR, based on an extension of SwinIR, originally designed for single image restoration, and the frequency domain knowledge obtained by the Fast Fourier Convolution (FFC). Specifically, to effectively gather global information, we modify the Residual Swin Transformer blocks (RSTBs) in SwinIR by explicitly incorporating the frequency domain knowledge using the FFC and employing the resulting residual Swin Fourier Transformer blocks (RSFTBs) for feature extraction. Besides, for the efficient and accurate fusion of stereo views, we propose a new cross-attention module referred to as RCAM, which achieves highly competitive performance while requiring less computational cost than the state-of-the-art cross-attention modules. Extensive experimental results and ablation studies demonstrate the effectiveness and efficiency of our proposed SwinFSR.

cs.CV↗

arXiv2023

TC-GAT: Graph Attention Network for Temporal Causality Discovery

The present study explores the intricacies of causal relationship extraction, a vital component in the pursuit of causality knowledge. Causality is frequently intertwined with temporal elements, as the progression from cause to effect is not instantaneous but rather ensconced in a temporal dimension. Thus, the extraction of temporal causality holds paramount significance in the field. In light of this, we propose a method for extracting causality from the text that integrates both temporal and causal relations, with a particular focus on the time aspect. To this end, we first compile a dataset that encompasses temporal relationships. Subsequently, we present a novel model, TC-GAT, which employs a graph attention mechanism to assign weights to the temporal relationships and leverages a causal knowledge graph to determine the adjacency matrix. Additionally, we implement an equilibrium mechanism to regulate the interplay between temporal and causal relations. Our experiments demonstrate that our proposed method significantly surpasses baseline models in the task of causality extraction.

cs.CL↗

arXiv2023

Weakly supervised segmentation with point annotations for histopathology images via contrast-based variational model

Image segmentation is a fundamental task in the field of imaging and vision. Supervised deep learning for segmentation has achieved unparalleled success when sufficient training data with annotated labels are available. However, annotation is known to be expensive to obtain, especially for histopathology images where the target regions are usually with high morphology variations and irregular shapes. Thus, weakly supervised learning with sparse annotations of points is promising to reduce the annotation workload. In this work, we propose a contrast-based variational model to generate segmentation results, which serve as reliable complementary supervision to train a deep segmentation model for histopathology images. The proposed method considers the common characteristics of target regions in histopathology images and can be trained in an end-to-end manner. It can generate more regionally consistent and smoother boundary segmentation, and is more robust to unlabeled `novel' regions. Experiments on two different histology datasets demonstrate its effectiveness and efficiency in comparison to previous models.

eess.IV↗

arXiv2023

Pac-HuBERT: Self-Supervised Music Source Separation via Primitive Auditory Clustering and Hidden-Unit BERT

In spite of the progress in music source separation research, the small amount of publicly-available clean source data remains a constant limiting factor for performance. Thus, recent advances in self-supervised learning present a largely-unexplored opportunity for improving separation models by leveraging unlabelled music data. In this paper, we propose a self-supervised learning framework for music source separation inspired by the HuBERT speech representation model. We first investigate the potential impact of the original HuBERT model by inserting an adapted version of it into the well-known Demucs V2 time-domain separation model architecture. We then propose a time-frequency-domain self-supervised model, Pac-HuBERT (for primitive auditory clustering HuBERT), that we later use in combination with a Res-U-Net decoder for source separation. Pac-HuBERT uses primitive auditory features of music as unsupervised clustering labels to initialize the self-supervised pretraining process using the Free Music Archive (FMA) dataset. The resulting framework achieves better source-to-distortion ratio (SDR) performance on the MusDB18 test set than the original Demucs V2 and Res-U-Net models. We further demonstrate that it can boost performance with small amounts of supervised data. Ultimately, our proposed framework is an effective solution to the challenge of limited clean source data for music source separation.

cs.SD↗

arXiv2023

Towards Solving Fuzzy Tasks with Human Feedback: A Retrospective of the MineRL BASALT 2022 Competition

To facilitate research in the direction of fine-tuning foundation models from human feedback, we held the MineRL BASALT Competition on Fine-Tuning from Human Feedback at NeurIPS 2022. The BASALT challenge asks teams to compete to develop algorithms to solve tasks with hard-to-specify reward functions in Minecraft. Through this competition, we aimed to promote the development of algorithms that use human feedback as channels to learn the desired behavior. We describe the competition and provide an overview of the top solutions. We conclude by discussing the impact of the competition and future directions for improvement.

cs.AI↗

arXiv2023

On the Maximal Independent Sets of $k$-mers with the Edit Distance

In computational biology, $k$-mers and edit distance are fundamental concepts. However, little is known about the metric space of all $k$-mers equipped with the edit distance. In this work, we explore the structure of the $k$-mer space by studying its maximal independent sets (MISs). An MIS is a sparse sketch of all $k$-mers with nice theoretical properties, and therefore admits critical applications in clustering, indexing, hashing, and sketching large-scale sequencing data, particularly those with high error-rates. Finding an MIS is a challenging problem, as the size of a $k$-mer space grows geometrically with respect to $k$. We propose three algorithms for this problem. The first and the most intuitive one uses a greedy strategy. The second method implements two techniques to avoid redundant comparisons by taking advantage of the locality-property of the $k$-mer space and the estimated bounds on the edit distance. The last algorithm avoids expensive calculations of the edit distance by translating the edit distance into the shortest path in a specifically designed graph. These algorithms are implemented and the calculated MISs of $k$-mer spaces and their statistical properties are reported and analyzed for $k$ up to 15. Source code is freely available at https://github.com/Shao-Group/kmerspace .

cs.DS↗

arXiv2023

Fast Multi-grid Methods for Minimizing Curvature Energy

The geometric high-order regularization methods such as mean curvature and Gaussian curvature, have been intensively studied during the last decades due to their abilities in preserving geometric properties including image edges, corners, and contrast. However, the dilemma between restoration quality and computational efficiency is an essential roadblock for high-order methods. In this paper, we propose fast multi-grid algorithms for minimizing both mean curvature and Gaussian curvature energy functionals without sacrificing accuracy for efficiency. Unlike the existing approaches based on operator splitting and the Augmented Lagrangian method (ALM), no artificial parameters are introduced in our formulation, which guarantees the robustness of the proposed algorithm. Meanwhile, we adopt the domain decomposition method to promote parallel computing and use the fine-to-coarse structure to accelerate convergence. Numerical experiments are presented on image denoising, CT, and MRI reconstruction problems to demonstrate the superiority of our method in preserving geometric structures and fine details. The proposed method is also shown effective in dealing with large-scale image processing problems by recovering an image of size $1024\times 1024$ within $40$s, while the ALM method requires around $200$s.

eess.IV↗

arXiv2023

An Interpretable MRI Reconstruction Network with Two-grid-cycle Correction and Geometric Prior Distillation

Although existing deep learning compressed-sensing-based Magnetic Resonance Imaging (CS-MRI) methods have achieved considerably impressive performance, explainability and generalizability continue to be challenging for such methods since the transition from mathematical analysis to network design not always natural enough, often most of them are not flexible enough to handle multi-sampling-ratio reconstruction assignments. {In this work, to tackle explainability and generalizability, we propose a unifying deep unfolding multi-sampling-ratio interpretable CS-MRI framework.} The combined approach offers more generalizability than previous works whereas deep learning gains explainability through a geometric prior module. Inspired by the multigrid algorithm, we first embed the CS-MRI-based optimization algorithm into correction-distillation scheme that consists of three ingredients: pre-relaxation module, correction module and geometric prior distillation module. Furthermore, we employ a condition module to learn adaptively step-length and noise level, which enables the proposed framework to jointly train multi-ratio tasks through a single model. { The proposed model not only compensates for the lost contextual information of reconstructed image which is refined from low frequency error in geometric characteristic k-space}, but also integrates the theoretical guarantee of model-based methods and the superior reconstruction performances of deep learning-based methods. Therefore, it can give us a novel perspective to design biomedical imaging networks. { Numerical experiments show that our framework outperforms state-of-the-art methods in terms of qualitative and quantitative evaluations.} {Our method achieves 3.18 dB improvement at low CS ratio 10\% and average 1.42 dB improvement over other comparison methods on brain dataset using Cartesian sampling mask.

eess.IV↗

arXiv2023

NVRadarNet: Real-Time Radar Obstacle and Free Space Detection for Autonomous Driving

Detecting obstacles is crucial for safe and efficient autonomous driving. To this end, we present NVRadarNet, a deep neural network (DNN) that detects dynamic obstacles and drivable free space using automotive RADAR sensors. The network utilizes temporally accumulated data from multiple RADAR sensors to detect dynamic obstacles and compute their orientation in a top-down bird's-eye view (BEV). The network also regresses drivable free space to detect unclassified obstacles. Our DNN is the first of its kind to utilize sparse RADAR signals in order to perform obstacle and free space detection in real time from RADAR data only. The network has been successfully used for perception on our autonomous vehicles in real self-driving scenarios. The network runs faster than real time on an embedded GPU and shows good generalization across geographic regions.

cs.CV↗

arXiv2023

Closed-Loop Transcription via Convolutional Sparse Coding

Autoencoding has achieved great empirical success as a framework for learning generative models for natural images. Autoencoders often use generic deep networks as the encoder or decoder, which are difficult to interpret, and the learned representations lack clear structure. In this work, we make the explicit assumption that the image distribution is generated from a multi-stage sparse deconvolution. The corresponding inverse map, which we use as an encoder, is a multi-stage convolution sparse coding (CSC), with each stage obtained from unrolling an optimization algorithm for solving the corresponding (convexified) sparse coding program. To avoid computational difficulties in minimizing distributional distance between the real and generated images, we utilize the recent closed-loop transcription (CTRL) framework that optimizes the rate reduction of the learned sparse representations. Conceptually, our method has high-level connections to score-matching methods such as diffusion models. Empirically, our framework demonstrates competitive performance on large-scale datasets, such as ImageNet-1K, compared to existing autoencoding and generative methods under fair conditions. Even with simpler networks and fewer computational resources, our method demonstrates high visual quality in regenerated images. More surprisingly, the learned autoencoder performs well on unseen datasets. Our method enjoys several side benefits, including more structured and interpretable representations, more stable convergence, and scalability to large datasets. Our method is arguably the first to demonstrate that a concatenation of multiple convolution sparse coding/decoding layers leads to an interpretable and effective autoencoder for modeling the distribution of large-scale natural image datasets.

cs.CV↗

arXiv2023

An Efficient Semi-Automated Scheme for Infrastructure LiDAR Annotation

Most existing perception systems rely on sensory data acquired from cameras, which perform poorly in low light and adverse weather conditions. To resolve this limitation, we have witnessed advanced LiDAR sensors become popular in perception tasks in autonomous driving applications. Nevertheless, their usage in traffic monitoring systems is less ubiquitous. We identify two significant obstacles in cost-effectively and efficiently developing such a LiDAR-based traffic monitoring system: (i) public LiDAR datasets are insufficient for supporting perception tasks in infrastructure systems, and (ii) 3D annotations on LiDAR point clouds are time-consuming and expensive. To fill this gap, we present an efficient semi-automated annotation tool that automatically annotates LiDAR sequences with tracking algorithms while offering a fully annotated infrastructure LiDAR dataset -- FLORIDA (Florida LiDAR-based Object Recognition and Intelligent Data Annotation) -- which will be made publicly available. Our advanced annotation tool seamlessly integrates multi-object tracking (MOT), single-object tracking (SOT), and suitable trajectory post-processing techniques. Specifically, we introduce a human-in-the-loop schema in which annotators recursively fix and refine annotations imperfectly predicted by our tool and incrementally add them to the training dataset to obtain better SOT and MOT models. By repeating the process, we significantly increase the overall annotation speed by three to four times and obtain better qualitative annotations than a state-of-the-art annotation tool. The human annotation experiments verify the effectiveness of our annotation tool. In addition, we provide detailed statistics and object detection evaluation results for our dataset in serving as a benchmark for perception tasks at traffic intersections.

cs.CV↗

arXiv2023

On the Stretch Factor of Polygonal Chains

Let $P=(p_1, p_2, \dots, p_n)$ be a polygonal chain in $\mathbb{R}^d$. The stretch factor of $P$ is the ratio between the total length of $P$ and the distance of its endpoints, $\sum_{i = 1}^{n-1} |p_i p_{i+1}|/|p_1 p_n|$. For a parameter $c \geq 1$, we call $P$ a $c$-chain if $|p_ip_j|+|p_jp_k| \leq c|p_ip_k|$, for every triple $(i,j,k)$, $1 \leq i 0$, there is a noncrossing $c$-chain that has stretch factor $Ω(n^{1/2-\varepsilon})$, for sufficiently large constant $c=c(\varepsilon)$; (ii) on the other hand, the stretch factor of a $c$-chain $P$ is $O\left(n^{1/2}\right)$, for every constant $c\geq 1$, regardless of whether $P$ is crossing or noncrossing; and (iii) we give a randomized algorithm that can determine, for a polygonal chain $P$ in $\mathbb{R}^2$ with $n$ vertices, the minimum $c\geq 1$ for which $P$ is a $c$-chain in $O\left(n^{2.5}\ \mathrm{polylog}\ n\right)$ expected time and $O(n\log n)$ space. These results generalize to $\mathbb{R}^d$. For every dimension $d\geq 2$ and every $\varepsilon>0$, we construct a noncrossing $c$-chain that has stretch factor $Ω\left(n^{(1-\varepsilon)(d-1)/d}\right)$; on the other hand, the stretch factor of any $c$-chain is $O\left((n-1)^{(d-1)/d}\right)$; for every $c>1$, we can test whether an $n$-vertex chain in $\mathbb{R}^d$ is a $c$-chain in $O\left(n^{3-1/d}\ \mathrm{polylog}\ n\right)$ expected time and $O(n\log n)$ space.

cs.CG↗

arXiv2023

Ultra-thin Epitaxial MgB2 on SiC: Substrate Surface Polarity Dependent Properties

High quality, ultrathin, superconducting films are required for advanced devices such as hot-electron bolometers, superconducting nanowire single photon detectors, and quantum applications. Using Hybrid Physical-Chemical Vapor Deposition (HPCVD), we show that MgB2 films as thin as 4 nm can be fabricated on the carbon terminated 6H-SiC (0001) surface with a superconducting transition temperature above 33K and a rms roughness of 0.7 nm. Remarkably, the film quality is a function of the SiC surface termination, with the C-terminated surface preferred to the Si-terminated surface. To understand the MgB2 thin film/ SiC substrate interactions giving rise to this difference, we characterized the interfacial structures using Rutherford backscattering spectroscopy/channeling, electron energy loss spectroscopy, and x-ray photoemission spectroscopy. The MgB2/SiC interface structure is complex and different for the two terminations. Both terminations incorporate substantial unintentional oxide layers influencing MgB2 growth and morphology, but with different extent, intermixing and interface chemistry. In this paper, we report measurements of transport, resistivity, and critical superconducting temperature of MgB2/SiC that are different for the two terminations, and link interfacial structure variations to observed differences. The result shows that the C face of SiC is a preferred substrate for the deposition of ultrathin superconducting MgB2 films.

cond-mat.supr-con↗

arXiv2022

Graph Contrastive Learning with Implicit Augmentations

Existing graph contrastive learning methods rely on augmentation techniques based on random perturbations (e.g., randomly adding or dropping edges and nodes). Nevertheless, altering certain edges or nodes can unexpectedly change the graph characteristics, and choosing the optimal perturbing ratio for each dataset requires onerous manual tuning. In this paper, we introduce Implicit Graph Contrastive Learning (iGCL), which utilizes augmentations in the latent space learned from a Variational Graph Auto-Encoder by reconstructing graph topological structure. Importantly, instead of explicitly sampling augmentations from latent distributions, we further propose an upper bound for the expected contrastive loss to improve the efficiency of our learning algorithm. Thus, graph semantics can be preserved within the augmentations in an intelligent way without arbitrary manual design or prior human knowledge. Experimental results on both graph-level and node-level tasks show that the proposed method achieves state-of-the-art performance compared to other benchmarks, where ablation studies in the end demonstrate the effectiveness of modules in iGCL.

cs.LG↗

arXiv2022

Gate-tunable negative refraction of mid-infrared polaritons

Negative refraction provides an attractive platform to manipulate mid-infrared and terahertz radiation for molecular sensing and thermal radiation applications. However, its implementation based on available metamaterials and plasmonic media presents challenges associated with optical losses, limited spatial confinement, and lack of active tunability in this spectral range. Here, we demonstrate gate-tunable negative refraction at mid-infrared frequencies using hybrid topological polaritons in van der Waals heterostructures with high spatial confinement. We experimentally visualize wide-angle negatively-refracted surface polaritons on α-MoO3 films partially decorated with graphene, undergoing planar nanoscale focusing down to 1.6% of the free-space wavelength. Our atomically thick heterostructures outperform conventional bulk materials by avoiding scattering losses at the refracting interface while enabling active tunability through electrical gating. We propose polaritonic negative refraction as a promising platform for infrared applications such as electrically tunable super-resolution imaging, nanoscale thermal manipulation, and molecular sensing.

physics.optics↗

arXiv2022

Global well-posedness of the 1d compressible Navier-Stokes system with rough data

In this paper, we study the global well-posedness problem for the 1d compressible Navier-Stokers system (cNSE) in gas dynamics with rough initial data. Frist, Liu- Yu (2022) established the global well-posedness theory for the 1d isentropic cNSE with initial velocity data in BV space. Then, it was extended to the 1d cNSE for the polytropic ideal gas with initial velocity and temperature data in BV space by Wang-Yu-Zhang (2022). We improve the global well-posedness result of Liu-Yu with initial velocity data in $W^{2γ,1}$ space; and of Wang-Yu-Zhang with initial velocity data in $ L^2\cap W^{2γ,1}$ space and initial data of temperature in $\dot W^{-\frac{2}{3},\frac{6}{5}}\cap \dot W^{2γ-1,1}$ for any $γ>0$ \textit{arbitrary small}. Our essential ideas are based on establishing various "end-point" smoothing estimates for the 1d parabolic equation.

math.AP↗