Search Results for author: Jingwei Zhang

Found 49 papers, 12 papers with code

Boosting Deep Neural Network Efficiency with Dual-Module Inference

no code implementations ICML 2020 Liu Liu, Lei Deng, Zhaodong Chen, yuke wang, Shuangchen Li, Jingwei Zhang, Yihua Yang, Zhenyu Gu, Yufei Ding, Yuan Xie

Using Deep Neural Networks (DNNs) in machine learning tasks is promising in delivering high-quality results but challenging to meet stringent latency requirements and energy constraints because of the memory-bound and the compute-bound execution pattern of DNNs.

MoreauPruner: Robust Pruning of Large Language Models against Weight Perturbations

1 code implementation11 Jun 2024 Zixiao Wang, Jingwei Zhang, Wenqian Zhao, Farzan Farnia, Bei Yu

Our numerical results suggest the robustness of MoreauPruner against weight perturbations, and indicate the MoreauPruner's successful accuracy-based scores in comparison to several existing pruning methods.

Diffusion-Refined VQA Annotations for Semi-Supervised Gaze Following

no code implementations4 Jun 2024 Qiaomu Miao, Alexandros Graikos, Jingwei Zhang, Sounak Mondal, Minh Hoai, Dimitris Samaras

We obtain the first prior using a large pretrained Visual Question Answering (VQA) model, where we compute Grad-CAM heatmaps by `prompting' the VQA model with a gaze following question.

Question Answering Visual Question Answering

An Empirical Study of Training State-of-the-Art LiDAR Segmentation Models

1 code implementation23 May 2024 Jiahao Sun, Chunmei Qing, Xiang Xu, Lingdong Kong, Youquan Liu, Li Li, Chenming Zhu, Jingwei Zhang, Zeqi Xiao, Runnan Chen, Tai Wang, Wenwei Zhang, Kai Chen

In the rapidly evolving field of autonomous driving, precise segmentation of LiDAR data is crucial for understanding complex 3D environments.

Autonomous Driving Benchmarking +3

Sparse Domain Transfer via Elastic Net Regularization

no code implementations13 May 2024 Jingwei Zhang, Farzan Farnia

Transportation of samples across different domains is a central task in several machine learning problems.

feature selection

Towards a Scalable Identification of Novel Modes in Generative Models

no code implementations4 May 2024 Jingwei Zhang, Mohammad Jalali, Cheuk Ting Li, Farzan Farnia

An interpretable comparison of generative models requires the identification of sample types produced more frequently by each of the involved models.

Integrative Graph-Transformer Framework for Histopathology Whole Slide Image Representation and Classification

no code implementations26 Mar 2024 Zhan Shi, Jingwei Zhang, Jun Kong, Fusheng Wang

In digital pathology, the multiple instance learning (MIL) strategy is widely used in the weakly supervised histopathology whole slide image (WSI) classification task where giga-pixel WSIs are only labeled at the slide level.

Multiple Instance Learning

An Interpretable Evaluation of Entropy-based Novelty of Generative Models

1 code implementation27 Feb 2024 Jingwei Zhang, Cheuk Ting Li, Farzan Farnia

In this work, we focus on the novelty assessment for multi-modal distributions and attempt to address the following differential clustering task: Given samples of a generative model $P_\mathcal{G}$ and a reference model $P_\mathrm{ref}$, how can we discover the sample types expressed by $P_\mathcal{G}$ more frequently than in $P_\mathrm{ref}$?

SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology

1 code implementation CVPR 2024 Saarthak Kapse, Pushpak Pati, Srijan Das, Jingwei Zhang, Chao Chen, Maria Vakalopoulou, Joel Saltz, Dimitris Samaras, Rajarsi R. Gupta, Prateek Prasanna

Introducing interpretability and reasoning into Multiple Instance Learning (MIL) methods for Whole Slide Image (WSI) analysis is challenging, given the complexity of gigapixel slides.

Multiple Instance Learning

Equivariant Data Augmentation for Generalization in Offline Reinforcement Learning

no code implementations14 Sep 2023 Cristina Pinneri, Sarah Bechtle, Markus Wulfmeier, Arunkumar Byravan, Jingwei Zhang, William F. Whitney, Martin Riedmiller

We present a novel approach to address the challenge of generalization in offline reinforcement learning (RL), where the agent learns from a fixed dataset without any additional interaction with the environment.

Data Augmentation Offline RL +2

SAM-Path: A Segment Anything Model for Semantic Segmentation in Digital Pathology

no code implementations12 Jul 2023 Jingwei Zhang, Ke Ma, Saarthak Kapse, Joel Saltz, Maria Vakalopoulou, Prateek Prasanna, Dimitris Samaras

On these two datasets, the proposed additional pathology foundation model further achieves a relative improvement of 5. 07% to 5. 12% in Dice score and 4. 50% to 8. 48% in IOU.

Instance Segmentation Segmentation +1

DiffFlow: A Unified SDE Framework for Score-Based Diffusion Models and Generative Adversarial Networks

no code implementations5 Jul 2023 Jingwei Zhang, Han Shi, Jincheng Yu, Enze Xie, Zhenguo Li

Generative models can be categorized into two types: explicit generative models that define explicit density forms and allow exact likelihood inference, such as score-based diffusion models (SDMs) and normalizing flows; implicit generative models that directly learn a transformation from the prior to the data distribution, such as generative adversarial nets (GANs).

Denoising

Prompt-MIL: Boosting Multi-Instance Learning Schemes via Task-specific Prompt Tuning

1 code implementation21 Mar 2023 Jingwei Zhang, Saarthak Kapse, Ke Ma, Prateek Prasanna, Joel Saltz, Maria Vakalopoulou, Dimitris Samaras

Compared to conventional full fine-tuning approaches, we fine-tune less than 1. 3% of the parameters, yet achieve a relative improvement of 1. 29%-13. 61% in accuracy and 3. 22%-27. 18% in AUROC and reduce GPU memory consumption by 38%-45% while training 21%-27% faster.

Leveraging Jumpy Models for Planning and Fast Learning in Robotic Domains

no code implementations24 Feb 2023 Jingwei Zhang, Jost Tobias Springenberg, Arunkumar Byravan, Leonard Hasenclever, Abbas Abdolmaleki, Dushyant Rao, Nicolas Heess, Martin Riedmiller

We conduct a set of experiments in the RGB-stacking environment, showing that planning with the learned skills and the associated model can enable zero-shot generalization to new tasks, and can further speed up training of policies via reinforcement learning.

reinforcement-learning Reinforcement Learning (RL) +1

MoreauGrad: Sparse and Robust Interpretation of Neural Networks via Moreau Envelope

1 code implementation ICCV 2023 Jingwei Zhang, Farzan Farnia

Explaining the predictions of deep neural nets has been a topic of great interest in the computer vision literature.

Clusterformer: Cluster-based Transformer for 3D Object Detection in Point Clouds

no code implementations ICCV 2023 Yu Pei, Xian Zhao, Hao Li, Jingyuan Ma, Jingwei Zhang, ShiLiang Pu

Attributed to the unstructured and sparse nature of point clouds, the transformer shows greater potential in point clouds data processing.

3D Object Detection Decoder +2

Local Learning on Transformers via Feature Reconstruction

no code implementations29 Dec 2022 Priyank Pathak, Jingwei Zhang, Dimitris Samaras

In this paper, we propose a new mechanism for each local module, where instead of reconstructing the entire image, we reconstruct its input features, generated from previous modules.

Precise Location Matching Improves Dense Contrastive Learning in Digital Pathology

1 code implementation23 Dec 2022 Jingwei Zhang, Saarthak Kapse, Ke Ma, Prateek Prasanna, Maria Vakalopoulou, Joel Saltz, Dimitris Samaras

Our method outperforms previous dense matching methods by up to 7. 2% in average precision for detection and 5. 6% in average precision for instance segmentation tasks.

Contrastive Learning Instance Segmentation +2

Gigapixel Whole-Slide Images Classification using Locally Supervised Learning

1 code implementation17 Jul 2022 Jingwei Zhang, Xin Zhang, Ke Ma, Rajarsi Gupta, Joel Saltz, Maria Vakalopoulou, Dimitris Samaras

Histopathology whole slide images (WSIs) play a very important role in clinical studies and serve as the gold standard for many cancer diagnoses.

Classification Multiple Instance Learning +1

Mean-Field Analysis of Two-Layer Neural Networks: Global Optimality with Linear Convergence Rates

no code implementations19 May 2022 Jingwei Zhang, Xunpeng Huang

We consider optimizing two-layer neural networks in the mean-field regime where the learning dynamics of network weights can be approximated by the evolution in the space of probability measures over the weight parameters associated with the neurons.

Memory-based Jitter: Improving Visual Recognition on Long-tailed Data with Diversity In Memory

no code implementations22 Aug 2020 Jialun Liu, Jingwei Zhang, Yi Yang, Wenhui Li, Chi Zhang, Yifan Sun

With slight modifications, MBJ is applicable for two fundamental visual recognition tasks, \emph{i. e.}, deep image classification and deep metric learning (on long-tailed data).

Data Augmentation General Classification +4

CycAs: Self-supervised Cycle Association for Learning Re-identifiable Descriptions

no code implementations ECCV 2020 Zhongdao Wang, Jingwei Zhang, Liang Zheng, Yixuan Liu, Yifan Sun, Ya-Li Li, Shengjin Wang

This paper proposes a self-supervised learning method for the person re-identification (re-ID) problem, where existing unsupervised methods usually rely on pseudo labels, such as those from video tracklets or clustering.

Clustering Multi-Object Tracking +2

Efficiency and Equity are Both Essential: A Generalized Traffic Signal Controller with Deep Reinforcement Learning

no code implementations9 Mar 2020 Shengchao Yan, Jingwei Zhang, Daniel Büscher, Wolfram Burgard

In this paper we present an approach to learning policies for signal controllers using deep reinforcement learning aiming for optimized traffic flow.

Generalization Bounds for Convolutional Neural Networks

no code implementations3 Oct 2019 Shan Lin, Jingwei Zhang

Convolutional neural networks (CNNs) have achieved breakthrough performances in a wide range of applications including image classification, semantic segmentation, and object detection.

Generalization Bounds Image Classification +3

Dual-module Inference for Efficient Recurrent Neural Networks

no code implementations25 Sep 2019 Liu Liu, Lei Deng, Shuangchen Li, Jingwei Zhang, Yihua Yang, Zhenyu Gu, Yufei Ding, Yuan Xie

Using Recurrent Neural Networks (RNNs) in sequence modeling tasks is promising in delivering high-quality results but challenging to meet stringent latency requirements because of the memory-bound execution pattern of RNNs.

Scheduled Intrinsic Drive: A Hierarchical Take on Intrinsically Motivated Exploration

no code implementations18 Mar 2019 Jingwei Zhang, Niklas Wetzel, Nicolai Dorka, Joschka Boedecker, Wolfram Burgard

Many state-of-the-art methods use intrinsic motivation to complement the sparse extrinsic reward signal, giving the agent more opportunities to receive feedback during exploration.

An Optimal Transport View on Generalization

no code implementations8 Nov 2018 Jingwei Zhang, Tongliang Liu, DaCheng Tao

We derive upper bounds on the generalization error of learning algorithms based on their \emph{algorithmic transport cost}: the expected Wasserstein distance between the output hypothesis and the output hypothesis conditioned on an input example.

Learning Theory

An Information-Theoretic View for Deep Learning

no code implementations24 Apr 2018 Jingwei Zhang, Tongliang Liu, DaCheng Tao

This upper bound shows that as the number of convolutional and pooling layers $L$ increases in the network, the expected generalization error will decrease exponentially to zero.

speech-recognition Speech Recognition

On the Rates of Convergence from Surrogate Risk Minimizers to the Bayes Optimal Classifier

no code implementations11 Feb 2018 Jingwei Zhang, Tongliang Liu, DaCheng Tao

We study the rates of convergence from empirical surrogate risk minimizers to the Bayes optimal classifier.

VR-Goggles for Robots: Real-to-sim Domain Adaptation for Visual Control

no code implementations1 Feb 2018 Jingwei Zhang, Lei Tai, Peng Yun, Yufeng Xiong, Ming Liu, Joschka Boedecker, Wolfram Burgard

In this paper, we deal with the reality gap from a novel perspective, targeting transferring Deep Reinforcement Learning (DRL) policies learned in simulated environments to the real-world domain for visual control tasks.

Domain Adaptation Style Transfer

Socially Compliant Navigation through Raw Depth Inputs with Generative Adversarial Imitation Learning

1 code implementation6 Oct 2017 Lei Tai, Jingwei Zhang, Ming Liu, Wolfram Burgard

Experiments show that our GAIL-based approach greatly improves the safety and efficiency of the behavior of mobile robots from pure behavior cloning.

Autonomous Vehicles Imitation Learning +1

Neural SLAM: Learning to Explore with External Memory

1 code implementation29 Jun 2017 Jingwei Zhang, Lei Tai, Ming Liu, Joschka Boedecker, Wolfram Burgard

We present an approach for agents to learn representations of a global map from sensor data, to aid their exploration in new environments.

Reinforcement Learning (RL) Simultaneous Localization and Mapping

A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation

1 code implementation21 Dec 2016 Lei Tai, Jingwei Zhang, Ming Liu, Joschka Boedecker, Wolfram Burgard

We carry out our discussions on the two main paradigms for learning control with deep networks: deep reinforcement learning and imitation learning.

Imitation Learning reinforcement-learning +1

Deep Reinforcement Learning with Successor Features for Navigation across Similar Environments

no code implementations16 Dec 2016 Jingwei Zhang, Jost Tobias Springenberg, Joschka Boedecker, Wolfram Burgard

We propose a successor feature based deep reinforcement learning algorithm that can learn to transfer knowledge from previously mastered navigation tasks to new problem instances.

reinforcement-learning Reinforcement Learning (RL) +1

Fast, Flexible Models for Discovering Topic Correlation across Weakly-Related Collections

1 code implementation EMNLP 2015 Jingwei Zhang, Aaron Gerow, Jaan Altosaar, James Evans, Richard Jean So

Weak topic correlation across document collections with different numbers of topics in individual collections presents challenges for existing cross-collection topic models.

Topic Models

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