arXiv subjects

Jingyu Li

Publications and source records attributed to Jingyu Li.

At least 91 records · Page 5Linked to original sources

arXiv2023

DDS3D: Dense Pseudo-Labels with Dynamic Threshold for Semi-Supervised 3D Object Detection

In this paper, we present a simple yet effective semi-supervised 3D object detector named DDS3D. Our main contributions have two-fold. On the one hand, different from previous works using Non-Maximal Suppression (NMS) or its variants for obtaining the sparse pseudo labels, we propose a dense pseudo-label generation strategy to get dense pseudo-labels, which can retain more potential supervision information for the student network. On the other hand, instead of traditional fixed thresholds, we propose a dynamic threshold manner to generate pseudo-labels, which can guarantee the quality and quantity of pseudo-labels during the whole training process. Benefiting from these two components, our DDS3D outperforms the state-of-the-art semi-supervised 3d object detection with mAP of 3.1% on the pedestrian and 2.1% on the cyclist under the same configuration of 1% samples. Extensive ablation studies on the KITTI dataset demonstrate the effectiveness of our DDS3D. The code and models will be made publicly available at https://github.com/hust-jy/DDS3D

cs.CV↗

arXiv2023

Private Data Valuation and Fair Payment in Data Marketplaces

Data valuation is an essential task in a data marketplace. It aims at fairly compensating data owners for their contribution. There is increasing recognition in the machine learning community that the Shapley value -- a foundational profit-sharing scheme in cooperative game theory -- has major potential to value data, because it uniquely satisfies basic properties for fair credit allocation and has been shown to be able to identify data sources that are useful or harmful to model performance. However, calculating the Shapley value requires accessing original data sources. It still remains an open question how to design a real-world data marketplace that takes advantage of the Shapley value-based data pricing while protecting privacy and allowing fair payments. In this paper, we propose the {\em first} prototype of a data marketplace that values data sources based on the Shapley value in a privacy-preserving manner and at the same time ensures fair payments. Our approach is enabled by a suite of innovations on both algorithm and system design. We firstly propose a Shapley value calculation algorithm that can be efficiently implemented via multiparty computation (MPC) circuits. The key idea is to learn a performance predictor that can directly predict model performance corresponding to an input dataset without performing actual training. We further optimize the MPC circuit design based on the structure of the performance predictor. We further incorporate fair payment into the MPC circuit to guarantee that the data that the buyer pays for is exactly the same as the one that has been valuated. Our experimental results show that the proposed new data valuation algorithm is as effective as the original expensive one. Furthermore, the customized MPC protocol is efficient and scalable.

cs.CR↗

arXiv2022

Global solutions to the Nernst-Planck-Euler system on bounded domain

We show that the Nernst-Planck-Euler system, which models ionic electrodiffusion in fluids, has global strong solutions for arbitrarily large data in the two dimensional bounded domains. The assumption on species is either there are two species or the diffusivities and the absolute values of ionic valences are the same if the species are arbitrarily many. In particular, the boundary conditions for the ions are allowed to be inhomogeneous. The proof is based on the energy estimates, integration along the characteristic line and the regularity theory of elliptic and parabolic equations.

math.AP↗

arXiv2022

Convolution-Based Channel-Frequency Attention for Text-Independent Speaker Verification

Deep convolutional neural networks (CNNs) have been applied to extracting speaker embeddings with significant success in speaker verification. Incorporating the attention mechanism has shown to be effective in improving the model performance. This paper presents an efficient two-dimensional convolution-based attention module, namely C2D-Att. The interaction between the convolution channel and frequency is involved in the attention calculation by lightweight convolution layers. This requires only a small number of parameters. Fine-grained attention weights are produced to represent channel and frequency-specific information. The weights are imposed on the input features to improve the representation ability for speaker modeling. The C2D-Att is integrated into a modified version of ResNet for speaker embedding extraction. Experiments are conducted on VoxCeleb datasets. The results show that C2DAtt is effective in generating discriminative attention maps and outperforms other attention methods. The proposed model shows robust performance with different scales of model size and achieves state-of-the-art results.

eess.AS↗

arXiv2022

Meta-Causal Feature Learning for Out-of-Distribution Generalization

Causal inference has become a powerful tool to handle the out-of-distribution (OOD) generalization problem, which aims to extract the invariant features. However, conventional methods apply causal learners from multiple data splits, which may incur biased representation learning from imbalanced data distributions and difficulty in invariant feature learning from heterogeneous sources. To address these issues, this paper presents a balanced meta-causal learner (BMCL), which includes a balanced task generation module (BTG) and a meta-causal feature learning module (MCFL). Specifically, the BTG module learns to generate balanced subsets by a self-learned partitioning algorithm with constraints on the proportions of sample classes and contexts. The MCFL module trains a meta-learner adapted to different distributions. Experiments conducted on NICO++ dataset verified that BMCL effectively identifies the class-invariant visual regions for classification and may serve as a general framework to improve the performance of the state-of-the-art methods.

cs.CV↗

arXiv2022

What Makes for Automatic Reconstruction of Pulmonary Segments

3D reconstruction of pulmonary segments plays an important role in surgical treatment planning of lung cancer, which facilitates preservation of pulmonary function and helps ensure low recurrence rates. However, automatic reconstruction of pulmonary segments remains unexplored in the era of deep learning. In this paper, we investigate what makes for automatic reconstruction of pulmonary segments. First and foremost, we formulate, clinically and geometrically, the anatomical definitions of pulmonary segments, and propose evaluation metrics adhering to these definitions. Second, we propose ImPulSe (Implicit Pulmonary Segment), a deep implicit surface model designed for pulmonary segment reconstruction. The automatic reconstruction of pulmonary segments by ImPulSe is accurate in metrics and visually appealing. Compared with canonical segmentation methods, ImPulSe outputs continuous predictions of arbitrary resolutions with higher training efficiency and fewer parameters. Lastly, we experiment with different network inputs to analyze what matters in the task of pulmonary segment reconstruction. Our code is available at https://github.com/M3DV/ImPulSe.

eess.IV↗

arXiv2022

Transport-Oriented Feature Aggregation for Speaker Embedding Learning

Pooling is needed to aggregate frame-level features into utterance-level representations for speaker modeling. Given the success of statistics-based pooling methods, we hypothesize that speaker characteristics are well represented in the statistical distribution over the pre-aggregation layer's output, and propose to use transport-oriented feature aggregation for deriving speaker embeddings. The aggregated representation encodes the geometric structure of the underlying feature distribution, which is expected to contain valuable speaker-specific information that may not be represented by the commonly used statistical measures like mean and variance. The original transport-oriented feature aggregation is also extended to a weighted-frame version to incorporate the attention mechanism. Experiments on speaker verification with the Voxceleb dataset show improvement over statistics pooling and its attentive variant.

eess.AS↗

arXiv2022

EDITnet: A Lightweight Network for Unsupervised Domain Adaptation in Speaker Verification

Performance degradation caused by language mismatch is a common problem when applying a speaker verification system on speech data in different languages. This paper proposes a domain transfer network, named EDITnet, to alleviate the language-mismatch problem on speaker embeddings without requiring speaker labels. The network leverages a conditional variational auto-encoder to transfer embeddings from the target domain into the source domain. A self-supervised learning strategy is imposed on the transferred embeddings so as to increase the cosine distance between embeddings from different speakers. In the training process of the EDITnet, the embedding extraction model is fixed without fine-tuning, which renders the training efficient and low-cost. Experiments on Voxceleb and CN-Celeb show that the embeddings transferred by EDITnet outperform the un-transferred ones by around 30% with the ECAPA-TDNN512. Performance improvement can also be achieved with other embedding extraction models, e.g., TDNN, SE-ResNet34.

eess.AS↗

arXiv2022

Learnable Frequency Filters for Speech Feature Extraction in Speaker Verification

Mel-scale spectrum features are used in various recognition and classification tasks on speech signals. There is no reason to expect that these features are optimal for all different tasks, including speaker verification (SV). This paper describes a learnable front-end feature extraction model. The model comprises a group of filters to transform the Fourier spectrum. Model parameters that define these filters are trained end-to-end and optimized specifically for the task of speaker verification. Compared to the standard Mel-scale filter-bank, the filters' bandwidths and center frequencies are adjustable. Experimental results show that applying the learnable acoustic front-end improves speaker verification performance over conventional Mel-scale spectrum features. Analysis on the learned filter parameters suggests that narrow-band information benefits the SV system performance. The proposed model achieves a good balance between performance and computation cost. In resource-constrained computation settings, the model significantly outperforms CNN-based learnable front-ends. The generalization ability of the proposed model is also demonstrated on different embedding extraction models and datasets.

eess.AS↗

arXiv2022

An Investigation on Applying Acoustic Feature Conversion to ASR of Adult and Child Speech

The performance of child speech recognition is generally less satisfactory compared to adult speech due to limited amount of training data. Significant performance degradation is expected when applying an automatic speech recognition (ASR) system trained on adult speech to child speech directly, as a result of domain mismatch. The present study is focused on adult-to-child acoustic feature conversion to alleviate this mismatch. Different acoustic feature conversion approaches, including deep neural network based and signal processing based, are investigated and compared under a fair experimental setting, in which converted acoustic features from the same amount of labeled adult speech are used to train the ASR models from scratch. Experimental results reveal that not all of the conversion methods lead to ASR performance gain. Specifically, as a classic unsupervised domain adaptation method, the statistic matching does not show an effectiveness. A disentanglement-based auto-encoder (DAE) conversion framework is found to be useful and the approach of F0 normalization achieves the best performance. It is noted that the F0 distribution of converted features is an important attribute to reflect the conversion quality, while utilizing an adult-child deep classification model to make judgment is shown to be inappropriate.

eess.AS↗

arXiv2022

Absorption bias: An ideal descriptor for radiation tolerance of nanocrystalline BCC metals

To evaluate the radiation tolerance of nanocrystalline (NC) materials, the damage effects of Fe and W as typical body-centered cubic (BCC) metals under uniform irradiation are studied by a sequential multi-scale modelling framework. An ideal descriptor, the absorption bias (the ratio of the absorption abilities of grain boundaries (GBs) to interstitials (I) and vacancies (V)), is proposed to characterize the radiation tolerance of materials with different grain sizes. Low absorption bias promotes defects annihilation through enhancing I-V recombination and optimally tuning its competition with GB absorption. Thus, the lower absorption bias, the higher anti-irradiation performance of NC BCC metals is. Furthermore, by comprehensively considering the mechanical property, thermal stability and radiation resistance, nano-crystals are recommended for Fe-based structural materials but coarse crystals for W-based plasma-facing materials. This work reevaluates the radiation resistance of NC materials, resulting in new strategies for designing structural materials of nuclear devices through manipulating grain sizes.

cond-mat.mtrl-sci↗

arXiv2021

Exploiting Pre-Trained ASR Models for Alzheimer's Disease Recognition Through Spontaneous Speech

Alzheimer's disease (AD) is a progressive neurodegenerative disease and recently attracts extensive attention worldwide. Speech technology is considered a promising solution for the early diagnosis of AD and has been enthusiastically studied. Most recent works concentrate on the use of advanced BERT-like classifiers for AD detection. Input to these classifiers are speech transcripts produced by automatic speech recognition (ASR) models. The major challenge is that the quality of transcription could degrade significantly under complex acoustic conditions in the real world. The detection performance, in consequence, is largely limited. This paper tackles the problem via tailoring and adapting pre-trained neural-network based ASR model for the downstream AD recognition task. Only bottom layers of the ASR model are retained. A simple fully-connected neural network is added on top of the tailored ASR model for classification. The heavy BERT classifier is discarded. The resulting model is light-weight and can be fine-tuned in an end-to-end manner for AD recognition. Our proposed approach takes only raw speech as input, and no extra transcription process is required. The linguistic information of speech is implicitly encoded in the tailored ASR model and contributes to boosting the performance. Experiments show that our proposed approach outperforms the best manual transcript-based RoBERTa by an absolute margin of 4.6% in terms of accuracy. Our best-performing models achieve the accuracy of 83.2% and 78.0% in the long-audio and short-audio competition tracks of the 2021 NCMMSC Alzheimer's Disease Recognition Challenge, respectively.

eess.AS↗

arXiv2021

Improving Text-Independent Speaker Verification with Auxiliary Speakers Using Graph

The paper presents a novel approach to refining similarity scores between input utterances for robust speaker verification. Given the embeddings from a pair of input utterances, a graph model is designed to incorporate additional information from a group of embeddings representing the so-called auxiliary speakers. The relations between the input utterances and the auxiliary speakers are represented by the edges and vertices in the graph. The similarity scores are refined by iteratively updating the values of the graph's vertices using an algorithm similar to the random walk algorithm on graphs. Through this updating process, the information of auxiliary speakers is involved in determining the relation between input utterances and hence contributing to the verification process. We propose to create a set of artificial embeddings through the model training process. Utilizing the generated embeddings as auxiliary speakers, no extra data are required for the graph model in the verification stage. The proposed model is trained in an end-to-end manner within the whole system. Experiments are carried out with the Voxceleb datasets. The results indicate that involving auxiliary speakers with graph is effective to improve speaker verification performance.

eess.AS↗

arXiv2021

Detection of Consonant Errors in Disordered Speech Based on Consonant-vowel Segment Embedding

Speech sound disorder (SSD) refers to a type of developmental disorder in young children who encounter persistent difficulties in producing certain speech sounds at the expected age. Consonant errors are the major indicator of SSD in clinical assessment. Previous studies on automatic assessment of SSD revealed that detection of speech errors concerning short and transitory consonants is less satisfactory. This paper investigates a neural network based approach to detecting consonant errors in disordered speech using consonant-vowel (CV) diphone segment in comparison to using consonant monophone segment. The underlying assumption is that the vowel part of a CV segment carries important information of co-articulation from the consonant. Speech embeddings are extracted from CV segments by a recurrent neural network model. The similarity scores between the embeddings of the test segment and the reference segments are computed to determine if the test segment is the expected consonant or not. Experimental results show that using CV segments achieves improved performance on detecting speech errors concerning those "difficult" consonants reported in the previous studies.

eess.AS↗

arXiv2020

Prediction of superconductivity and topological aspects in single-layer $β$-Bi$_{2}$Pd

Topological superconductors, characterized by topologically nontrivial states residing in a superconducting gap, are a recently discovered class of materials having Majorana Fermions. The interplay of superconductivity and topological states give rise to opportunities for achieving such topological superconductors in condensed matter systems. Up to now, several single-material topological superconductors in this form have been theoretically predicted and experimentally confirmed. Here, using the first-principles calculations, we study the superconducting single-layer $β$-Bi$_{2}$Pd. The electronic density of states near Fermi level of this monolayer are dominated by the Bi-p and Pd-d orbitals, forming a two-band Fermi surface with multi-class sheets. The presence of soft phonon bands, in cooperation with the electron susceptibility, account for electron-phonon superconductivity of single-layer $β$-Bi$_{2}$Pd. With the centrosymmetric structure, single-layer $β$-Bi$_{2}$Pd possesses a continuous gap over the whole Brillouin zone and topological Dirac-like states at its one-dimensional boundary. The present findings would lead to the expectation of one-dimensional topological superconductivity and Majorana bound states in monolayer candidate of $β$-Bi$_{2}$Pd with intrinsic full-gap superconductivity.

cond-mat.mtrl-sci↗

arXiv2020

Text-Independent Speaker Verification with Dual Attention Network

This paper presents a novel design of attention model for text-independent speaker verification. The model takes a pair of input utterances and generates an utterance-level embedding to represent speaker-specific characteristics in each utterance. The input utterances are expected to have highly similar embeddings if they are from the same speaker. The proposed attention model consists of a self-attention module and a mutual attention module, which jointly contributes to the generation of the utterance-level embedding. The self-attention weights are computed from the utterance itself while the mutual-attention weights are computed with the involvement of the other utterance in the input pairs. As a result, each utterance is represented by a self-attention weighted embedding and a mutual-attention weighted embedding. The similarity between the embeddings is measured by a cosine distance score and a binary classifier output score. The whole model, named Dual Attention Network, is trained end-to-end on Voxceleb database. The evaluation results on Voxceleb 1 test set show that the Dual Attention Network significantly outperforms the baseline systems. The best result yields an equal error rate of 1:6%.

eess.AS↗

arXiv2019

Convergence to traveling waves of a singular PDE-ODE hybrid chemotaxis system in the half space

This paper is concerned with the asymptotic stability of the initial-boundary value problem of a singular PDE-ODE hybrid chemotaxis system in the half space $\R_+=[0, \infty)$. We show that when the non-zero flux boundary condition at $x=0$ is prescribed and the initial data are suitably chosen, the solution of the initial-boundary value problem converges, as time tend to infinity, to a shifted traveling wavefront restricted in the half space $[0,\infty)$ where the wave profile and speed are uniquely selected by the boundary flux data. The results are proved by a Cole-Hopf type transformation and weighted energy estimates along with the technique of taking {\color{black} the} anti-derivative.

math.AP↗

arXiv2019

Boundary spike-layer solutions of the singular Keller-Segel system: existence and stability

We exploit the existence and nonlinear stability of boundary spike/layer solutions of the Keller-Segel system with logarithmic singular sensitivity in the half space, where the physical zero-flux and Dirichlet boundary conditions are prescribed. We first prove that, under above boundary conditions, the Keller-Segel system admits a unique boundary spike-layer steady state where the first solution component (bacterial density) of the system concentrates at the boundary as a Dirac mass and the second solution component (chemical concentration) forms a boundary layer profile near the boundary as the chemical diffusion coefficient tends to zero. Then we show that this boundary spike-layer steady state is asymptotically nonlinearly stable under appropriate perturbations. As far as we know, this is the first result obtained on the global well-posedness of the singular Keller-Segel system with nonlinear consumption rate. We introduce a novel strategy of relegating the singularity, via a Cole-Hopf type transformation, to a nonlinear nonlocality which is resolved by the technique of "taking antiderivatives", i.e. working at the level of the distribution function. Then, we carefully choose weight functions to prove our main results by suitable weighted energy estimates with Hardy's inequality that fully captures the dissipative structure of the system.

math.AP↗