Search Results for author: Mao Ye

Found 58 papers, 21 papers with code

Network Pruning by Greedy Subnetwork Selection

no code implementations ICML 2020 Mao Ye, Chengyue Gong, Lizhen Nie, Denny Zhou, Adam Klivans, Qiang Liu

Theoretically, we show that the small networks pruned using our method achieve provably lower loss than small networks trained from scratch with the same size.

Network Pruning

Triple-domain Feature Learning with Frequency-aware Memory Enhancement for Moving Infrared Small Target Detection

1 code implementation11 Jun 2024 Weiwei Duan, Luping Ji, Shengjia Chen, Sicheng Zhu, Mao Ye

To extend target feature learning, we propose a new Triple-domain Strategy (Tridos) with the frequency-aware memory enhancement on the spatial-temporal domain.

Proxy Denoising for Source-Free Domain Adaptation

1 code implementation3 Jun 2024 Song Tang, Wenxin Su, Mao Ye, Jianwei Zhang, Xiatian Zhu

This is grounded on a novel proxy confidence theory by modeling elegantly the domain adaption effect of the proxy's divergence against the domain-invariant space.

Denoising Source-Free Domain Adaptation

Unified Source-Free Domain Adaptation

1 code implementation12 Mar 2024 Song Tang, Wenxin Su, Mao Ye, Jianwei Zhang, Xiatian Zhu

To tackle this unified SFDA problem, we propose a novel approach called Latent Causal Factors Discovery (LCFD).

Language Modelling Source-Free Domain Adaptation +1

Source-Free Domain Adaptation with Frozen Multimodal Foundation Model

1 code implementation CVPR 2024 Song Tang, Wenxin Su, Mao Ye, Xiatian Zhu

We find that directly applying the ViL model to the target domain in a zero-shot fashion is unsatisfactory, as it is not specialized for this particular task but largely generic.

Source-Free Domain Adaptation

Spatial-Temporal Transformer based Video Compression Framework

no code implementations21 Sep 2023 Yanbo Gao, Wenjia Huang, Shuai Li, Hui Yuan, Mao Ye, Siwei Ma

Similar as the traditional video coding, LVC inherits motion estimation/compensation, residual coding and other modules, all of which are implemented with neural networks (NNs).

Motion Estimation Video Compression

MSMix:An Interpolation-Based Text Data Augmentation Method Manifold Swap Mixup

no code implementations31 May 2023 Mao Ye, Haitao Wang, Zheqian Chen

To solve the problem of poor performance of deep neural network models due to insufficient data, a simple yet effective interpolation-based data augmentation method is proposed: MSMix (Manifold Swap Mixup).

Data Augmentation Intent Detection

Independent Feature Decomposition and Instance Alignment for Unsupervised Domain Adaptation

1 code implementation IJCAI 2023 Qichen He, Siying Xiao, Mao Ye, Xiatian Zhu, Ferrante Neri and Dongde Hou

Existing Unsupervised Domain Adaptation (UDA) methods typically attempt to perform knowledge transfer in a domain-invariant space explicitly or implicitly.

Transfer Learning Unsupervised Domain Adaptation

Efficient Transformer-based 3D Object Detection with Dynamic Token Halting

no code implementations ICCV 2023 Mao Ye, Gregory P. Meyer, Yuning Chai, Qiang Liu

Although halting a token is a non-differentiable operation, our method allows for differentiable end-to-end learning by leveraging an equivalent differentiable forward-pass.

3D Object Detection Autonomous Vehicles +1

Homeomorphism Alignment for Unsupervised Domain Adaptation

1 code implementation ICCV 2023 Lihua Zhou, Mao Ye, Xiatian Zhu, Siying Xiao, Xu-Qian Fan, Ferrante Neri

With distribution alignment, it is challenging to acquire a common space which maintains fully the discriminative structure of both domains.

Pseudo Label Self-Supervised Learning +1

BOME! Bilevel Optimization Made Easy: A Simple First-Order Approach

no code implementations19 Sep 2022 Mao Ye, Bo Liu, Stephen Wright, Peter Stone, Qiang Liu

Bilevel optimization (BO) is useful for solving a variety of important machine learning problems including but not limited to hyperparameter optimization, meta-learning, continual learning, and reinforcement learning.

Bilevel Optimization Continual Learning +3

Future Gradient Descent for Adapting the Temporal Shifting Data Distribution in Online Recommendation Systems

no code implementations2 Sep 2022 Mao Ye, Ruichen Jiang, Haoxiang Wang, Dhruv Choudhary, Xiaocong Du, Bhargav Bhushanam, Aryan Mokhtari, Arun Kejariwal, Qiang Liu

One of the key challenges of learning an online recommendation model is the temporal domain shift, which causes the mismatch between the training and testing data distribution and hence domain generalization error.

Domain Generalization Recommendation Systems

First Hitting Diffusion Models for Generating Manifold, Graph and Categorical Data

no code implementations2 Sep 2022 Mao Ye, Lemeng Wu, Qiang Liu

We propose a family of First Hitting Diffusion Models (FHDM), deep generative models that generate data with a diffusion process that terminates at a random first hitting time.

Diffusion-based Molecule Generation with Informative Prior Bridges

no code implementations2 Sep 2022 Lemeng Wu, Chengyue Gong, Xingchao Liu, Mao Ye, Qiang Liu

AI-based molecule generation provides a promising approach to a large area of biomedical sciences and engineering, such as antibody design, hydrolase engineering, or vaccine development.

3D Generation Point Cloud Generation

Let us Build Bridges: Understanding and Extending Diffusion Generative Models

no code implementations31 Aug 2022 Xingchao Liu, Lemeng Wu, Mao Ye, Qiang Liu

Diffusion-based generative models have achieved promising results recently, but raise an array of open questions in terms of conceptual understanding, theoretical analysis, algorithm improvement and extensions to discrete, structured, non-Euclidean domains.

Imputation

Multi-Class 3D Object Detection with Single-Class Supervision

no code implementations11 May 2022 Mao Ye, Chenxi Liu, Maoqing Yao, Weiyue Wang, Zhaoqi Leng, Charles R. Qi, Dragomir Anguelov

While multi-class 3D detectors are needed in many robotics applications, training them with fully labeled datasets can be expensive in labeling cost.

3D Object Detection Object +1

Source-Free Object Detection by Learning To Overlook Domain Style

1 code implementation CVPR 2022 Shuaifeng Li, Mao Ye, Xiatian Zhu, Lihua Zhou, Lin Xiong

This approach suffers from both unsatisfactory accuracy of pseudo labels due to the presence of domain shift and limited use of target domain training data.

object-detection Object Detection

argmax centroid

no code implementations NeurIPS 2021 Chengyue Gong, Mao Ye, Qiang Liu

We propose a general method to construct centroid approximation for the distribution of maximum points of a random function (a. k. a.

Domain Adaptation Few-Shot Image Classification +2

Pareto Navigation Gradient Descent: a First-Order Algorithm for Optimization in Pareto Set

no code implementations17 Oct 2021 Mao Ye, Qiang Liu

The notion of the Pareto set allows us to focus on the set of (often infinite number of) models that cannot be strictly improved.

Multi-Task Learning

Centroid Approximation for Bootstrap: Improving Particle Quality at Inference

no code implementations17 Oct 2021 Mao Ye, Qiang Liu

In this work, we propose an efficient method to explicitly \emph{optimize} a small set of high quality ``centroid'' points to better approximate the ideal bootstrap distribution.

Uncertainty Quantification

Pareto Navigation Gradient Descent: a First Order Algorithm for Optimization in Pareto Set

no code implementations29 Sep 2021 Mao Ye, Qiang Liu

The notion of the Pareto set allows us to focus on the set of (often infinite number of) models that cannot be strictly improved.

Multi-Task Learning

MaxUp: Lightweight Adversarial Training With Data Augmentation Improves Neural Network Training

no code implementations CVPR 2021 Chengyue Gong, Tongzheng Ren, Mao Ye, Qiang Liu

The idea is to generate a set of augmented data with some random perturbations or transforms, and minimize the maximum, or worst case loss over the augmented data.

Data Augmentation Image Classification +1

VCNet and Functional Targeted Regularization For Learning Causal Effects of Continuous Treatments

1 code implementation14 Mar 2021 Lizhen Nie, Mao Ye, Qiang Liu, Dan Nicolae

Motivated by the rising abundance of observational data with continuous treatments, we investigate the problem of estimating the average dose-response curve (ADRF).

QoS-aware Link Scheduling Strategy for Data Transmission in SDVN

no code implementations1 Feb 2021 Yong Zhang, Mao Ye, Lin Guan

The original contributions of this paper are summarized as follows: (1) Model the packets collision probability of broadcast or NACK transmission in VANET with the combination theory and investigate the potential influence of miss my packets (MMP) problem.

Networking and Internet Architecture

Varying Coefficient Neural Network with Functional Targeted Regularization for Estimating Continuous Treatment Effects

no code implementations ICLR 2021 Lizhen Nie, Mao Ye, Qiang Liu, Dan Nicolae

With the rising abundance of observational data with continuous treatments, we investigate the problem of estimating average dose-response curve (ADRF).

Greedy Optimization Provably Wins the Lottery: Logarithmic Number of Winning Tickets is Enough

1 code implementation NeurIPS 2020 Mao Ye, Lemeng Wu, Qiang Liu

Despite the great success of deep learning, recent works show that large deep neural networks are often highly redundant and can be significantly reduced in size.

Adaptive Dense-to-Sparse Paradigm for Pruning Online Recommendation System with Non-Stationary Data

no code implementations16 Oct 2020 Mao Ye, Dhruv Choudhary, Jiecao Yu, Ellie Wen, Zeliang Chen, Jiyan Yang, Jongsoo Park, Qiang Liu, Arun Kejariwal

To the best of our knowledge, this is the first work to provide in-depth analysis and discussion of applying pruning to online recommendation systems with non-stationary data distribution.

Recommendation Systems

Go Wide, Then Narrow: Efficient Training of Deep Thin Networks

no code implementations ICML 2020 Denny Zhou, Mao Ye, Chen Chen, Tianjian Meng, Mingxing Tan, Xiaodan Song, Quoc Le, Qiang Liu, Dale Schuurmans

This is achieved by layerwise imitation, that is, forcing the thin network to mimic the intermediate outputs of the wide network from layer to layer.

Computational Efficiency Model Compression

SAFER: A Structure-free Approach for Certified Robustness to Adversarial Word Substitutions

1 code implementation ACL 2020 Mao Ye, Chengyue Gong, Qiang Liu

For security reasons, it is of critical importance to develop models with certified robustness that can provably guarantee that the prediction is can not be altered by any possible synonymous word substitution.

text-classification Text Classification

Unsupervised Feature Selection via Multi-step Markov Transition Probability

no code implementations29 May 2020 Yan Min, Mao Ye, Liang Tian, Yulin Jian, Ce Zhu, Shangming Yang

Our main contributions are a novel feature section approach which uses multi-step transition probability to characterize the data structure, and three algorithms proposed from the positive and negative aspects for keeping data structure.

Dimensionality Reduction feature selection +1

Learning Various Length Dependence by Dual Recurrent Neural Networks

no code implementations28 May 2020 Chenpeng Zhang, Shuai Li, Mao Ye, Ce Zhu, Xue Li

Many variants of RNN have been proposed to solve the gradient problems of training RNNs and process long sequences.

Disentanglement Then Reconstruction: Learning Compact Features for Unsupervised Domain Adaptation

no code implementations28 May 2020 Lihua Zhou, Mao Ye, Xinpeng Li, Ce Zhu, Yiguang Liu, Xue Li

By this reconstructor, we can construct prototypes for the original features using class prototypes and domain prototypes correspondingly.

Disentanglement Unsupervised Domain Adaptation

Steepest Descent Neural Architecture Optimization: Escaping Local Optimum with Signed Neural Splitting

no code implementations23 Mar 2020 Lemeng Wu, Mao Ye, Qi Lei, Jason D. Lee, Qiang Liu

Recently, Liu et al.[19] proposed a splitting steepest descent (S2D) method that jointly optimizes the neural parameters and architectures based on progressively growing network structures by splitting neurons into multiple copies in a steepest descent fashion.

Good Subnetworks Provably Exist: Pruning via Greedy Forward Selection

1 code implementation3 Mar 2020 Mao Ye, Chengyue Gong, Lizhen Nie, Denny Zhou, Adam Klivans, Qiang Liu

This differs from the existing methods based on backward elimination, which remove redundant neurons from the large network.

Network Pruning

Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework

no code implementations NeurIPS 2020 Dinghuai Zhang, Mao Ye, Chengyue Gong, Zhanxing Zhu, Qiang Liu

Randomized classifiers have been shown to provide a promising approach for achieving certified robustness against adversarial attacks in deep learning.

Stein Self-Repulsive Dynamics: Benefits From Past Samples

1 code implementation NeurIPS 2020 Mao Ye, Tongzheng Ren, Qiang Liu

Our idea is to introduce Stein variational gradient as a repulsive force to push the samples of Langevin dynamics away from the past trajectories.

MaxUp: A Simple Way to Improve Generalization of Neural Network Training

1 code implementation20 Feb 2020 Chengyue Gong, Tongzheng Ren, Mao Ye, Qiang Liu

The idea is to generate a set of augmented data with some random perturbations or transforms and minimize the maximum, or worst case loss over the augmented data.

Few-Shot Image Classification General Classification +1

Post-training Quantization with Multiple Points: Mixed Precision without Mixed Precision

no code implementations20 Feb 2020 Xingchao Liu, Mao Ye, Dengyong Zhou, Qiang Liu

We propose multipoint quantization, a quantization method that approximates a full-precision weight vector using a linear combination of multiple vectors of low-bit numbers; this is in contrast to typical quantization methods that approximate each weight using a single low precision number.

object-detection Object Detection +1

Extended Stochastic Gradient MCMC for Large-Scale Bayesian Variable Selection

1 code implementation7 Feb 2020 Qifan Song, Yan Sun, Mao Ye, Faming Liang

Stochastic gradient Markov chain Monte Carlo (MCMC) algorithms have received much attention in Bayesian computing for big data problems, but they are only applicable to a small class of problems for which the parameter space has a fixed dimension and the log-posterior density is differentiable with respect to the parameters.

Variable Selection

Distribution-Aware Coordinate Representation for Human Pose Estimation

6 code implementations CVPR 2020 Feng Zhang, Xiatian Zhu, Hanbin Dai, Mao Ye, Ce Zhu

Interestingly, we found that the process of decoding the predicted heatmaps into the final joint coordinates in the original image space is surprisingly significant for human pose estimation performance, which nevertheless was not recognised before.

Ranked #2 on Multi-Person Pose Estimation on MS COCO (using extra training data)

Keypoint Detection Multi-Person Pose Estimation

Strain engineering of epitaxial oxide heterostructures beyond substrate limitations

no code implementations3 May 2019 Xiong Deng, Chao Chen, Deyang Chen, Xiangbin Cai, Xiaozhe Yin, Chao Xu, Fei Sun, Caiwen Li, Yan Li, Han Xu, Mao Ye, Guo Tian, Zhen Fan, Zhipeng Hou, Minghui Qin, Yu Chen, Zhenlin Luo, Xubing Lu, Guofu Zhou, Lang Chen, Ning Wang, Ye Zhu, Xingsen Gao, Jun-Ming Liu

The limitation of commercially available single-crystal substrates and the lack of continuous strain tunability preclude the ability to take full advantage of strain engineering for further exploring novel properties and exhaustively studying fundamental physics in complex oxides.

Materials Science

Fast Human Pose Estimation

1 code implementation CVPR 2019 Feng Zhang, Xiatian Zhu, Mao Ye

In this work, we investigate the under-studied but practically critical pose model efficiency problem.

Pose Estimation

Stein Neural Sampler

1 code implementation8 Oct 2018 Tianyang Hu, Zixiang Chen, Hanxi Sun, Jincheng Bai, Mao Ye, Guang Cheng

We propose two novel samplers to generate high-quality samples from a given (un-normalized) probability density.

Variable Selection via Penalized Neural Network: a Drop-Out-One Loss Approach

no code implementations ICML 2018 Mao Ye, Yan Sun

We propose a variable selection method for high dimensional regression models, which allows for complex, nonlinear, and high-order interactions among variables.

regression Variable Selection

Do Convolutional Neural Networks Learn Class Hierarchy?

no code implementations17 Oct 2017 Bilal Alsallakh, Amin Jourabloo, Mao Ye, Xiaoming Liu, Liu Ren

We present visual-analytics methods to reveal and analyze this hierarchy of similar classes in relation with CNN-internal data.

Image Classification

3D Reconstruction in the Presence of Glasses by Acoustic and Stereo Fusion

no code implementations CVPR 2015 Mao Ye, Yu Zhang, Ruigang Yang, Dinesh Manocha

We present a novel sensor fusion algorithm that first segments the depth map into different categories such as opaque/transparent/infinity (e. g., too far to measure) and then updates the depth map based on the segmentation outcome.

3D Reconstruction Sensor Fusion +1

Quality Dynamic Human Body Modeling Using a Single Low-cost Depth Camera

no code implementations CVPR 2014 Qing Zhang, Bo Fu, Mao Ye, Ruigang Yang

In this paper we present a novel autonomous pipeline to build a personalized parametric model (pose-driven avatar) using a single depth sensor.

Real-time Simultaneous Pose and Shape Estimation for Articulated Objects Using a Single Depth Camera

no code implementations CVPR 2014 Mao Ye, Ruigang Yang

In this paper we present a novel real-time algorithm for simultaneous pose and shape estimation for articulated objects, such as human beings and animals.

Pose Estimation

Data-driven Flower Petal Modeling with Botany Priors

no code implementations CVPR 2014 Chenxi Zhang, Mao Ye, Bo Fu, Ruigang Yang

Each segmented petal is then fitted with a scale-invariant morphable petal shape model, which is constructed from individually scanned exemplar petals.

Exploring Social Influence for Recommendation - A Probabilistic Generative Model Approach

no code implementations4 Sep 2011 Mao Ye, Xingjie Liu, Wang-Chien Lee

The experimental results also confirm that our social influence based group recommendation algorithm outperforms the state-of-the-art algorithms for group recommendation.

Collaborative Filtering

Cannot find the paper you are looking for? You can Submit a new open access paper.