arXiv subjects

Ang Li

Publications and source records attributed to Ang Li.

At least 451 records · Page 25Linked to original sources

arXiv2022

Prediction of Depression Severity Based on the Prosodic and Semantic Features with Bidirectional LSTM and Time Distributed CNN

Depression is increasingly impacting individuals both physically and psychologically worldwide. It has become a global major public health problem and attracts attention from various research fields. Traditionally, the diagnosis of depression is formulated through semi-structured interviews and supplementary questionnaires, which makes the diagnosis heavily relying on physicians experience and is subject to bias. Mental health monitoring and cloud-based remote diagnosis can be implemented through an automated depression diagnosis system. In this article, we propose an attention-based multimodality speech and text representation for depression prediction. Our model is trained to estimate the depression severity of participants using the Distress Analysis Interview Corpus-Wizard of Oz (DAIC-WOZ) dataset. For the audio modality, we use the collaborative voice analysis repository (COVAREP) features provided by the dataset and employ a Bidirectional Long Short-Term Memory Network (Bi-LSTM) followed by a Time-distributed Convolutional Neural Network (T-CNN). For the text modality, we use global vectors for word representation (GloVe) to perform word embeddings and the embeddings are fed into the Bi-LSTM network. Results show that both audio and text models perform well on the depression severity estimation task, with best sequence level F1 score of 0.9870 and patient-level F1 score of 0.9074 for the audio model over five classes (healthy, mild, moderate, moderately severe, and severe), as well as sequence level F1 score of 0.9709 and patient-level F1 score of 0.9245 for the text model over five classes. Results are similar for the multimodality fused model, with the highest F1 score of 0.9580 on the patient-level depression detection task over five classes. Experiments show statistically significant improvements over previous works.

cs.HC↗

arXiv2022

AEVA: Black-box Backdoor Detection Using Adversarial Extreme Value Analysis

Deep neural networks (DNNs) are proved to be vulnerable against backdoor attacks. A backdoor is often embedded in the target DNNs through injecting a backdoor trigger into training examples, which can cause the target DNNs misclassify an input attached with the backdoor trigger. Existing backdoor detection methods often require the access to the original poisoned training data, the parameters of the target DNNs, or the predictive confidence for each given input, which are impractical in many real-world applications, e.g., on-device deployed DNNs. We address the black-box hard-label backdoor detection problem where the DNN is fully black-box and only its final output label is accessible. We approach this problem from the optimization perspective and show that the objective of backdoor detection is bounded by an adversarial objective. Further theoretical and empirical studies reveal that this adversarial objective leads to a solution with highly skewed distribution; a singularity is often observed in the adversarial map of a backdoor-infected example, which we call the adversarial singularity phenomenon. Based on this observation, we propose the adversarial extreme value analysis(AEVA) to detect backdoors in black-box neural networks. AEVA is based on an extreme value analysis of the adversarial map, computed from the monte-carlo gradient estimation. Evidenced by extensive experiments across multiple popular tasks and backdoor attacks, our approach is shown effective in detecting backdoor attacks under the black-box hard-label scenarios.

cs.LG↗

arXiv2022

Bit-GraphBLAS: Bit-Level Optimizations of Matrix-Centric Graph Processing on GPU

In a general graph data structure like an adjacency matrix, when edges are homogeneous, the connectivity of two nodes can be sufficiently represented using a single bit. This insight has, however, not yet been adequately exploited by the existing matrix-centric graph processing frameworks. This work fills the void by systematically exploring the bit-level representation of graphs and the corresponding optimizations to the graph operations. It proposes a two-level representation named Bit-Block Compressed Sparse Row (B2SR) and presents a series of optimizations to the graph operations on B2SR by leveraging the intrinsics of modern GPUs. Evaluations on NVIDIA Pascal and Volta GPUs show that the optimizations bring up to $40\times$ and $6555\times$ for essential GraphBLAS kernels SpMV and SpGEMM, respectively, making GraphBLAS-based BFS accelerate up to $433\times$, SSSP, PR, and CC up to $35\times$, and TC up to $52\times$.

cs.DC↗

arXiv2022

Efimov resonance position near a narrow Feshbach resonance in $^6$Li-$^{133}$Cs mixture

In the vicinity of a narrow Feshbach resonances Efimov features are expected to be characterized by the resonance's properties rather than the van der Waals length of the interatomic potential. Although this theoretical prediction is well-established by now, it still lacks experimental confirmation. Here, we apply our recently developed three-channel model [Yudkin and Khaykovich, Phys. Rev. A 103, 063303 (2021)] to the experimental result obtained in a mass-imbalanced $^6$Li-$^{133}$Cs mixture in the vicinity of the narrowest resonance explored to date [Johansen at. al. Nat. Phys. 13, 731 (2017)]. We confirm that the observed position of the Efimov resonance is dictated mainly by the resonance physics while the influence of the van der Waals tail of the interatomic potential is minor. We show that the resonance position is strongly influenced by the presence of another Feshbach resonance which significantly alters the effective background scattering length at the narrow resonance position.

cond-mat.quant-gas↗

arXiv2022

Learning and Fast Adaptation for Grid Emergency Control via Deep Meta Reinforcement Learning

As power systems are undergoing a significant transformation with more uncertainties, less inertia and closer to operation limits, there is increasing risk of large outages. Thus, there is an imperative need to enhance grid emergency control to maintain system reliability and security. Towards this end, great progress has been made in developing deep reinforcement learning (DRL) based grid control solutions in recent years. However, existing DRL-based solutions have two main limitations: 1) they cannot handle well with a wide range of grid operation conditions, system parameters, and contingencies; 2) they generally lack the ability to fast adapt to new grid operation conditions, system parameters, and contingencies, limiting their applicability for real-world applications. In this paper, we mitigate these limitations by developing a novel deep meta reinforcement learning (DMRL) algorithm. The DMRL combines the meta strategy optimization together with DRL, and trains policies modulated by a latent space that can quickly adapt to new scenarios. We test the developed DMRL algorithm on the IEEE 300-bus system. We demonstrate fast adaptation of the meta-trained DRL polices with latent variables to new operating conditions and scenarios using the proposed method and achieve superior performance compared to the state-of-the-art DRL and model predictive control (MPC) methods.

cs.LG↗

arXiv2022

SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-Training for Spatial-Aware Visual Representations

Pre-training has become a standard paradigm in many computer vision tasks. However, most of the methods are generally designed on the RGB image domain. Due to the discrepancy between the two-dimensional image plane and the three-dimensional space, such pre-trained models fail to perceive spatial information and serve as sub-optimal solutions for 3D-related tasks. To bridge this gap, we aim to learn a spatial-aware visual representation that can describe the three-dimensional space and is more suitable and effective for these tasks. To leverage point clouds, which are much more superior in providing spatial information compared to images, we propose a simple yet effective 2D Image and 3D Point cloud Unsupervised pre-training strategy, called SimIPU. Specifically, we develop a multi-modal contrastive learning framework that consists of an intra-modal spatial perception module to learn a spatial-aware representation from point clouds and an inter-modal feature interaction module to transfer the capability of perceiving spatial information from the point cloud encoder to the image encoder, respectively. Positive pairs for contrastive losses are established by the matching algorithm and the projection matrix. The whole framework is trained in an unsupervised end-to-end fashion. To the best of our knowledge, this is the first study to explore contrastive learning pre-training strategies for outdoor multi-modal datasets, containing paired camera images and LIDAR point clouds. Codes and models are available at https://github.com/zhyever/SimIPU.

cs.CV↗

arXiv2021

M2DGR: A Multi-sensor and Multi-scenario SLAM Dataset for Ground Robots

We introduce M2DGR: a novel large-scale dataset collected by a ground robot with a full sensor-suite including six fish-eye and one sky-pointing RGB cameras, an infrared camera, an event camera, a Visual-Inertial Sensor (VI-sensor), an inertial measurement unit (IMU), a LiDAR, a consumer-grade Global Navigation Satellite System (GNSS) receiver and a GNSS-IMU navigation system with real-time kinematic (RTK) signals. All those sensors were well-calibrated and synchronized, and their data were recorded simultaneously. The ground truth trajectories were obtained by the motion capture device, a laser 3D tracker, and an RTK receiver. The dataset comprises 36 sequences (about 1TB) captured in diverse scenarios including both indoor and outdoor environments. We evaluate state-of-the-art SLAM algorithms on M2DGR. Results show that existing solutions perform poorly in some scenarios. For the benefit of the research community, we make the dataset and tools public. The webpage of our project is https://github.com/SJTU-ViSYS/M2DGR.

cs.RO↗

arXiv2021

A special Higgs challenge: Measuring the mass and production cross section with ultimate precision at FCC-ee

The FCC-ee offers powerful opportunities to determine the Higgs boson parameters, exploiting over $10^6$ ${\rm e^+e^- \to ZH}$ events and almost $10^5$ ${\rm WW \to H}$ events at centre-of-mass energies around 240 and 365 GeV. This essay spotlights the important measurements of the ZH production cross section and of the Higgs boson mass. The measurement of the total ZH cross section is an essential input to the absolute determination of the HZZ coupling -- a "standard candle" that can be used by all other measurements, including those made at hadron colliders -- at the per-mil level. A combination of the measured cross sections at the two different centre-of-mass energies further provides the first evidence for the trilinear Higgs self-coupling, and possibly its first observation if the cross-section measurement can be made accurate enough. The determination of the Higgs boson mass with a precision significantly better than the Higgs boson width (4.1 MeV in the Standard Model) is a prerequisite to either constrain or measure the electron Yukawa coupling via direct ${\rm e^+e^- \to H}$ production at $\sqrt{s} = 125$ GeV. Approaching the statistical limit of 0.1% and $\mathcal{O}(1)$ MeV on the ZH cross section and the Higgs boson mass, respectively, sets highly demanding requirements on accelerator operation (ZH threshold scan, centre-of-mass energy measurement), detector design (lepton momentum resolution, hadronic final state reconstruction performance), theoretical calculations, and analysis techniques (efficiency and purity optimization with modern tools, constrained kinematic fits, control of systematic uncertainties). These challenges are examined in turn in this essay.

hep-ex↗

arXiv2021

EQC : Ensembled Quantum Computing for Variational Quantum Algorithms

Variational quantum algorithm (VQA), which is comprised of a classical optimizer and a parameterized quantum circuit, emerges as one of the most promising approaches for harvesting the power of quantum computers in the noisy intermediate scale quantum (NISQ) era. However, the deployment of VQAs on contemporary NISQ devices often faces considerable system and time-dependant noise and prohibitively slow training speeds. On the other hand, the expensive supporting resources and infrastructure make quantum computers extremely keen on high utilization. In this paper, we propose a virtualized way of building up a quantum backend for variational quantum algorithms: rather than relying on a single physical device which tends to introduce temporal-dependant device-specific noise with worsening performance as time-since-calibration grows, we propose to constitute a quantum ensemble, which dynamically distributes quantum tasks asynchronously across a set of physical devices, and adjusting the ensemble configuration with respect to machine status. In addition to reduced machine-dependant noise, the ensemble can provide significant speedups for VQA training. With this idea, we build a novel VQA training framework called EQC that comprises: (i) a system architecture for asynchronous parallel VQA cooperative training; (ii) an analytic model for assessing the quality of the returned VQA gradient over a particular device concerning its architecture, transpilation, and runtime conditions; (iii) a weighting mechanism to adjust the quantum ensemble's computational contribution according to the systems' current performance. Evaluations comprising 500K circuit evaluations across 10 IBMQ devices using a VQE and a QAOA applications demonstrate that EQC can attain error rates close to the most performant device of the ensemble, while boosting the training speed by 10.5x on average (up to 86x and at least 5.2x).

quant-ph↗

arXiv2021

Effects of dark matter on the nuclear and neutron star matter

We study the dark matter effects on the nuclear matter parameters characterising the equation of states of super dense neutron-rich nucleonic-matter. The observables of the nuclear matter, i.e. incompressibility, symmetry energy and its higher-order derivatives in the presence dark matter for symmetric and asymmetric nuclear matter are analysed with the help of an extended relativistic mean-field model. The calculations are also extended to beta-stable matter to explore the properties of the neutron star. We analyse the dark matter effects on symmetric nuclear matter, pure neutron matter and neutron star using NL3, G3 and IOPB-I forces. The binding energy per particle and pressure are calculated with and without considering the dark matter interaction with the nuclear matter systems. The influences of dark matter are also analysed on the symmetry energy and its different coefficients. The incompressibility and the skewness parameters are affected considerably due to the presence of dark matter in the nuclear matter medium. We extend the calculations to the neutron star and find its mass, radius and the moment of inertia for static and rotating neutron star with and without dark matter contribution. The mass of the rotating neutron star is considerably changing due to rapid rotation with the frequency in the mass-shedding limit. The effects of dark matter are found to be important for some of the nuclear matter parameters, which are crucial for the properties of astrophysical objects.

nucl-th↗

arXiv2021

On the DLM/FD methods for simulating neutrally buoyant swimmer motion in non-Newtonian shear thinning fluids

In this article we discuss the generalization of a Lagrange multiplier based fictitious domain (DLM/FD) method to simulating the motion of neutrally buoyant particles of non-symmetric shape in non-Newtonian shear thinning fluids. Numerical solutions of steady Poiseuille flow of non-Newtonian shear thinning fluids are compared with the exact solutions in a two-dimensional channel. Concerning a self-propelled swimmer formed by two disks, the effect of shear thinning makes the swimmer moving faster and decreases the critical Reynolds number (for the moving direction changing to the opposite one) when decreasing the value of the power index in the Carreau-Bird model.

physics.flu-dyn↗

arXiv2021

APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores

Over the years, accelerating neural networks with quantization has been widely studied. Unfortunately, prior efforts with diverse precisions (e.g., 1-bit weights and 2-bit activations) are usually restricted by limited precision support on GPUs (e.g., int1 and int4). To break such restrictions, we introduce the first Arbitrary Precision Neural Network framework (APNN-TC) to fully exploit quantization benefits on Ampere GPU Tensor Cores. Specifically, APNN-TC first incorporates a novel emulation algorithm to support arbitrary short bit-width computation with int1 compute primitives and XOR/AND Boolean operations. Second, APNN-TC integrates arbitrary precision layer designs to efficiently map our emulation algorithm to Tensor Cores with novel batching strategies and specialized memory organization. Third, APNN-TC embodies a novel arbitrary precision NN design to minimize memory access across layers and further improve performance. Extensive evaluations show that APNN-TC can achieve significant speedup over CUTLASS kernels and various NN models, such as ResNet and VGG.

cs.DC↗

arXiv2021

One Pass ImageNet

We present the One Pass ImageNet (OPIN) problem, which aims to study the effectiveness of deep learning in a streaming setting. ImageNet is a widely known benchmark dataset that has helped drive and evaluate recent advancements in deep learning. Typically, deep learning methods are trained on static data that the models have random access to, using multiple passes over the dataset with a random shuffle at each epoch of training. Such data access assumption does not hold in many real-world scenarios where massive data is collected from a stream and storing and accessing all the data becomes impractical due to storage costs and privacy concerns. For OPIN, we treat the ImageNet data as arriving sequentially, and there is limited memory budget to store a small subset of the data. We observe that training a deep network in a single pass with the same training settings used for multi-epoch training results in a huge drop in prediction accuracy. We show that the performance gap can be significantly decreased by paying a small memory cost and utilizing techniques developed for continual learning, despite the fact that OPIN differs from typical continual problem settings. We propose using OPIN to study resource-efficient deep learning.

cs.LG↗

arXiv2021

FL-WBC: Enhancing Robustness against Model Poisoning Attacks in Federated Learning from a Client Perspective

Federated learning (FL) is a popular distributed learning framework that trains a global model through iterative communications between a central server and edge devices. Recent works have demonstrated that FL is vulnerable to model poisoning attacks. Several server-based defense approaches (e.g. robust aggregation), have been proposed to mitigate such attacks. However, we empirically show that under extremely strong attacks, these defensive methods fail to guarantee the robustness of FL. More importantly, we observe that as long as the global model is polluted, the impact of attacks on the global model will remain in subsequent rounds even if there are no subsequent attacks. In this work, we propose a client-based defense, named White Blood Cell for Federated Learning (FL-WBC), which can mitigate model poisoning attacks that have already polluted the global model. The key idea of FL-WBC is to identify the parameter space where long-lasting attack effect on parameters resides and perturb that space during local training. Furthermore, we derive a certified robustness guarantee against model poisoning attacks and a convergence guarantee to FedAvg after applying our FL-WBC. We conduct experiments on FasionMNIST and CIFAR10 to evaluate the defense against state-of-the-art model poisoning attacks. The results demonstrate that our method can effectively mitigate model poisoning attack impact on the global model within 5 communication rounds with nearly no accuracy drop under both IID and Non-IID settings. Our defense is also complementary to existing server-based robust aggregation approaches and can further improve the robustness of FL under extremely strong attacks.

cs.LG↗

arXiv2021

Revisiting the post-glitch relaxation of the 2000 Vela glitch with the neutron star equation of states in the Brueckner and relativistic Brueckner theories

We revisit the short-term post-glitch relaxation of the Vela 2000 glitch in the simple two-component model of pulsar glitch by making use of the latest realistic equations of states from the microscopic Brueckner and the relativistic Brueckner theories for neutron stars, which can reconcile with the available astrophysical constraints. We show that to fit both the glitch size and the post-glitch jumps in frequency derivatives approximately one minute after the glitch, the mass of the Vela pulsar is necessarily small, and there may be demands for a stiff equation of state (which results in a typical stellar radius larger than $\sim\negthickspace12.5\,\rm km$) and a strong suppression of the pairing gap in the nuclear medium. We discuss the implications of this result on the understanding of pulsar glitches.

astro-ph.HE↗

arXiv2021

Characterizing User Susceptibility to COVID-19 Misinformation on Twitter

Though significant efforts such as removing false claims and promoting reliable sources have been increased to combat COVID-19 "misinfodemic", it remains an unsolved societal challenge if lacking a proper understanding of susceptible online users, i.e., those who are likely to be attracted by, believe and spread misinformation. This study attempts to answer {\it who} constitutes the population vulnerable to the online misinformation in the pandemic, and what are the robust features and short-term behavior signals that distinguish susceptible users from others. Using a 6-month longitudinal user panel on Twitter collected from a geopolitically diverse network-stratified samples in the US, we distinguish different types of users, ranging from social bots to humans with various level of engagement with COVID-related misinformation. We then identify users' online features and situational predictors that correlate with their susceptibility to COVID-19 misinformation. This work brings unique contributions: First, contrary to the prior studies on bot influence, our analysis shows that social bots' contribution to misinformation sharing was surprisingly low, and human-like users' misinformation behaviors exhibit heterogeneity and temporal variability. While the sharing of misinformation was highly concentrated, the risk of occasionally sharing misinformation for average users remained alarmingly high. Second, our findings highlight the political sensitivity activeness and responsiveness to emotionally-charged content among susceptible users. Third, we demonstrate a feasible solution to efficiently predict users' transient susceptibility solely based on their short-term news consumption and exposure from their networks. Our work has an implication in designing effective intervention mechanism to mitigate the misinformation dissipation.

cs.SI↗

arXiv2021

G-CoS: GNN-Accelerator Co-Search Towards Both Better Accuracy and Efficiency

Graph Neural Networks (GNNs) have emerged as the state-of-the-art (SOTA) method for graph-based learning tasks. However, it still remains prohibitively challenging to inference GNNs over large graph datasets, limiting their application to large-scale real-world tasks. While end-to-end jointly optimizing GNNs and their accelerators is promising in boosting GNNs' inference efficiency and expediting the design process, it is still underexplored due to the vast and distinct design spaces of GNNs and their accelerators. In this work, we propose G-CoS, a GNN and accelerator co-search framework that can automatically search for matched GNN structures and accelerators to maximize both task accuracy and acceleration efficiency. Specifically, GCoS integrates two major enabling components: (1) a generic GNN accelerator search space which is applicable to various GNN structures and (2) a one-shot GNN and accelerator co-search algorithm that enables simultaneous and efficient search for optimal GNN structures and their matched accelerators. To the best of our knowledge, G-CoS is the first co-search framework for GNNs and their accelerators. Extensive experiments and ablation studies show that the GNNs and accelerators generated by G-CoS consistently outperform SOTA GNNs and GNN accelerators in terms of both task accuracy and hardware efficiency, while only requiring a few hours for the end-to-end generation of the best matched GNNs and their accelerators.

cs.AR↗

arXiv2021

Unit Selection with Causal Diagram

The unit selection problem aims to identify a set of individuals who are most likely to exhibit a desired mode of behavior, for example, selecting individuals who would respond one way if encouraged and a different way if not encouraged. Using a combination of experimental and observational data, Li and Pearl derived tight bounds on the "benefit function" - the payoff/cost associated with selecting an individual with given characteristics. This paper shows that these bounds can be narrowed significantly (enough to change decisions) when structural information is available in the form of a causal model. We address the problem of estimating the benefit function using observational and experimental data when specific graphical criteria are assumed to hold.

cs.AI↗