Jinghui Chen

Assistant Professor, Ph.D.

Penn State University

Email: jzc5917 [at] psu [dot] edu

About

I am an Assistant Professor in the College of Information Sciences and Technology at Penn State University. I received my Ph.D. in the Department of Computer Science, University of California, Los Angeles (UCLA) working with Prof. Quanquan Gu in 2021. I received my B.E. in the Department of Electrical Engineering and Information Science at the University of Science and Technology of China in 2015.

My research interests broadly include the theory and applications in different aspects of machine learning (machine learning robustness, machine learning efficiency, adversarial machine learning, etc.)

Openings: I’m looking for highly motivated students (including PhDs, Masters, undergraduates), and interns to join my group. If you’re interested in joining my lab, please fill and see instructions in the following form (feel free to skip optional questions).

News

Publications

Full publications on Google Scholar.
E indicates authors with equal contribution. # indicates students supervised.

Do Imperceptible Perturbations Really Prevent Unauthorized Data Usage in Diffusion-based Image Generation Systems?

Bochuan Cao#, Changjiang Li, Ting Wang, Jinyuan Jia, Bo Li and Jinghui Chen

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models

Ziyi Yin, Muchao Ye, Tianrong Zhang#, Tianyu Du, Jinguo Zhu, Han Liu, Jinghui Chen, Ting Wang and Fenglong Ma

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

A3FL: Adversarially Adaptive Backdoor Attacks to Federated Learning

Hangfan Zhang, Jinyuan Jia, Jinghui Chen, Lu Lin and Dinghao Wu

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

Defending Pre-trained Language Models as Few-shot Learners against Backdoor Attacks

Tianyu Du, Zhaohan Xi, Changjiang Li, Ren Pang, Shouling Ji, Jinghui Chen, Fenglong Ma and Ting Wang

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

UniT: A Unified Look at Certified Robust Training against Text Adversarial Perturbation

Muchao Ye, Ziyi Yin, Tianrong Zhang#, Tianyu Du, Jinghui Chen, Ting Wang and Fenglong Ma

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

RoCourseNet: Robust Training of a Prediction Aware Recourse Model

Hangzhi Guo, Feiran Jia, Jinghui Chen, Anna Squicciarini and Amulya Yadav

in Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM), Birmingham, UK, 2023.

A short version of this paper also appears on ICML 2022 Workshop on Adversarial Machine Learning Frontiers.

PAT: Geometry-Aware Hard-Label Black-Box Adversarial Attacks on Text

Muchao Ye, Jinghui Chen, Chenglin Miao, Han Liu, Ting Wang and Fenglong Ma

in Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), San Diego, CA, USA, 2023.

Benign Overfitting in Adversarially Robust Linear Classification

Jinghui ChenE, Yuan CaoE, and Quanquan Gu

in Proceedings of the 39th Conference on Uncertainty in Artificial Intelligence (UAI), Pittsburgh, PA, USA, 2023.

Graph Contrastive Backdoor Attacks

Hangfan Zhang, Jinghui Chen, Lu Lin, Jinyuan Jia and Dinghao Wu

in Proceedings of the 40th International Conference on Machine Learning (ICML), Hawaii, USA, 2023.

Multiple Models for Outbreak Decision Support in the Face of Uncertainty

Katriona Shea, ..., Jinghui Chen, ..., Michael C. Runge.

in Proceedings of the National Academy of Sciences (PNAS), 2023.

Do Language Models Plagiarize?

Lee, Jooyoung, Thai Le, Jinghui Chen, and Dongwon Lee

in Proceedings of the ACM Web Conference (WWW), Austin, Texas, USA, 2023

Spectral Augmentation for Self-Supervised Learning on Graphs

Lu Lin, Jinghui Chen, Hongning Wang

in Proceedings of the 11th International Conference on Learning Representations (ICLR), Kigali Rwanda, 2023.

A short version of this paper also appears on NeurIPS 2022 Workshop on New Frontiers in Graph Learning.

On the Vulnerability of Backdoor Defenses for Federated Learning

Pei Fang# and Jinghui Chen

in Proceedings of the 37th Conference on Artificial Intelligence (AAAI), Washington DC, USA, 2023.

A short version of this paper also appears on International Workshop on Federated Learning: Recent Advances and New Challenges in Conjunction with NeurIPS 2022.

One-shot Neural Backdoor Erasing via Adversarial Weight Masking

Shuwen Chai# and Jinghui Chen

in Proceedings of the 36th Advances in Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA, 2022.

How Powerful is Implicit Denoising in Graph Neural Networks

Songtao Liu, Zhitao Ying, Hanze Dong, Lu Lin, Jinghui Chen and Dinghao Wu

NeurIPS 2022 Workshop on New Frontiers in Graph Learning.

The United States COVID-19 Forecast Hub dataset

Estee Y Cramer, ..., Jinghui Chen, ..., Nicholas G. Reich

Scientific Data, 9(1), pp.1-15., 2022.

LeapAttack: Hard-Label Adversarial Attack on Text via Gradient-Based Optimization

Muchao Ye, Jinghui Chen, Chenglin Miao, Ting Wang and Fenglong Ma

in Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Washington DC, USA, 2022.

Communication-Efficient Adaptive Federated Learning

Yujia Wang# , Lu Lin and Jinghui Chen

in Proceedings of the 39th International Conference on Machine Learning (ICML), Baltimore, Maryland, USA, 2022.

Evaluation of individual and ensemble probabilistic forecasts of COVID-19 mortality in the US

Estee Y Cramer, ..., Jinghui Chen, ..., Nicholas G. Reich

in Proceedings of the National Academy of Sciences (PNAS), 2022.

Learnability Lock: Authorized Learnability Control Through Adversarial Invertible Transformations

Weiqi Peng# and Jinghui Chen

in Proceedings of the 10th International Conference on Learning Representations (ICLR), Virtual, 2022.

Communication-Compressed Adaptive Gradient Method for Distributed Nonconvex Optimization

Yujia Wang# , Lu Lin and Jinghui Chen

in Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS), Virtual, 2022.

A short version of this paper also appears on International Workshop on Trustable, Verifiable and Auditable Federated Learning in Conjunction with AAAI 2022 (FL-AAAI-22).

Efficient Robust Training via Backward Smoothing

Jinghui Chen, Yu Cheng, Zhe Gan, Quanquan Gu and Jingjing Liu

In Proceedings of the 36th AAAI Conference on Artificial Intelligence (AAAI), Vancouver, BC, Canada, 2022.

Do Wider Neural Networks Really Help Adversarial Robustness?

Boxi WuE, Jinghui ChenE, Deng Cai, Xiaofei He and Quanquan Gu

in Proceedings of the 35th Advances in Neural Information Processing Systems (NeurIPS), Virtual, 2021.

Benign Overfitting in Adversarially Robust Linear Classification

Jinghui ChenE, Yuan CaoE and Quanquan Gu

ICML 2021 Workshop on Overparameterization: Pitfalls and Opportunities.

Epidemic Model Guided Machine Learning for COVID-19 Forecasts in the United States

Difan Zou, Lingxiao Wang, Pan Xu, Jinghui Chen, Weitong Zhang and Quanquan Gu

ICLR 2021 Workshop on Machine Learning for Preventingand Combating Pandemics.

On the Convergence of Adaptive Gradient Methods for Nonconvex Optimization

Dongruo ZhouE, Jinghui ChenE, Yuan CaoE, Yiqi Tang, Ziyan Yang, and Quanquan Gu

NeurIPS 2020 Workshop on Optimization for Machine Learning.

Ensemble Forecasts of Coronavirus Disease 2019 (COVID-19) in the U.S

COVID-19 Forecast Hub Consortium, Jinghui Chen.

medRxiv:2020.08.19.20177493, 2020.

RayS: A Ray Searching Method for Hard-label Adversarial Attack

Jinghui Chen and Quanquan Gu

in Proceedings of the 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), San Diego, CA, USA 2020.

A short version of this paper also appears on ICML 2020 Workshop on Uncertainty & Robustness in Deep Learning and ECCV 2020 Workshop on Adversarial Robustness in the Real World.

Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks

Jinghui Chen, Dongruo Zhou, Yiqi Tang, Ziyan Yang, Yuan Cao and Quanquan Gu

in Proceedings of 29th International Joint Conference on Artificial Intelligence (IJCAI), Yokohama, Japan, 2020.

Understanding the Intrinsic Robustness of Image Distributions using Conditional Generative Models

Xiao ZhangE, Jinghui ChenE, Quanquan Gu and David Evans

in Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS), Palermo, Sicily, Italy, 2020.

A Frank-Wolfe Framework for Efficient and Effective Adversarial Attacks

Jinghui Chen, Dongruo Zhou, Jinfeng Yi and Quanquan Gu

in Proceedings of the 34th Conference on Artificial Intelligence (AAAI), New York, New York, USA, 2020.

Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimization

Pan XuE, Jinghui ChenE, Difan Zou and Quanquan Gu

in Proceedings of the 32nd Advances in Neural Information Processing Systems (NIPS), Montréal, Canada, 2018.

Covariate Adjusted Precision Matrix Estimation via Nonconvex Optimization

Jinghui Chen, Pan Xu, Lingxiao Wang, Jian Ma and Quanquan Gu

in Proceedings of the 35th International Conference on Machine Learning (ICML), Stockholm, Sweden, 2018.

Fast Newton Hard Thresholding Pursuit for Sparsity Constrained Nonconvex Optimization

Jinghui Chen and Quanquan Gu

in Proceedings of the 23rd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Halifax, Nova Scotia, Canada, 2017.

Robust Wirtinger Flow for Phase Retrieval with Arbitrary Corruption

Jinghui Chen, Lingxiao Wang, Xiao Zhang, and Quanquan Gu

arXiv:1704.06256, 2017.

Outlier Detection with Autoencoder Ensembles

Jinghui Chen, Saket Sathe, Charu Aggarwal, and Deepak Turaga

in Proceedings of the 17th SIAM International Conference on Data Mining (SDM), Houston, Texas, USA, 2017.

Stochastic Block Coordinate Gradient Descent for Sparsity Constrained Optimization

Jinghui Chen and Quanquan Gu

in Proceedings of the 32th International Conference on Uncertainty in Artificial Intelligence (UAI), New York, USA, 2016.

Improved threshold Selection by using Calibrated Probabilities for Random Forest Classifiers

Florian Baumann, Jinghui Chen, Karsten Vogt and Bodo Rosenhahn

in Proceedings of the 12th Conference on Computer and Robot Vision (CRV), Halifax, Nova Scotia, Canada, 2015.

Do Imperceptible Perturbations Really Prevent Unauthorized Data Usage in Diffusion-based Image Generation Systems?

Bochuan Cao#, Changjiang Li, Ting Wang, Jinyuan Jia, Bo Li and Jinghui Chen

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models

Ziyi Yin, Muchao Ye, Tianrong Zhang#, Tianyu Du, Jinguo Zhu, Han Liu, Jinghui Chen, Ting Wang and Fenglong Ma

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

A3FL: Adversarially Adaptive Backdoor Attacks to Federated Learning

Hangfan Zhang, Jinyuan Jia, Jinghui Chen, Lu Lin and Dinghao Wu

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

Defending Pre-trained Language Models as Few-shot Learners against Backdoor Attacks

Tianyu Du, Zhaohan Xi, Changjiang Li, Ren Pang, Shouling Ji, Jinghui Chen, Fenglong Ma and Ting Wang

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

UniT: A Unified Look at Certified Robust Training against Text Adversarial Perturbation

Muchao Ye, Ziyi Yin, Tianrong Zhang#, Tianyu Du, Jinghui Chen, Ting Wang and Fenglong Ma

in Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2023.

RoCourseNet: Robust Training of a Prediction Aware Recourse Model

Hangzhi Guo, Feiran Jia, Jinghui Chen, Anna Squicciarini and Amulya Yadav

in Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM), Birmingham, UK, 2023.

A short version of this paper also appears on ICML 2022 Workshop on Adversarial Machine Learning Frontiers.

PAT: Geometry-Aware Hard-Label Black-Box Adversarial Attacks on Text

Muchao Ye, Jinghui Chen, Chenglin Miao, Han Liu, Ting Wang and Fenglong Ma

in Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), San Diego, CA, USA, 2023.

Benign Overfitting in Adversarially Robust Linear Classification

Jinghui ChenE, Yuan CaoE, and Quanquan Gu

in Proceedings of the 39th Conference on Uncertainty in Artificial Intelligence (UAI), Pittsburgh, PA, USA, 2023.

Graph Contrastive Backdoor Attacks

Hangfan Zhang, Jinghui Chen, Lu Lin, Jinyuan Jia and Dinghao Wu

in Proceedings of the 40th International Conference on Machine Learning (ICML), Hawaii, USA, 2023.

Do Language Models Plagiarize?

Lee, Jooyoung, Thai Le, Jinghui Chen, and Dongwon Lee

in Proceedings of the Web Conference (WWW), Austin, Texas, USA, 2023

Spectral Augmentation for Self-Supervised Learning on Graphs

Lu Lin, Jinghui Chen, Hongning Wang

in Proceedings of the 11th International Conference on Learning Representations (ICLR), Kigali Rwanda, 2023.

A short version of this paper also appears on NeurIPS 2022 Workshop on New Frontiers in Graph Learning.

On the Vulnerability of Backdoor Defenses for Federated Learning

Pei Fang# and Jinghui Chen

in Proceedings of the 37th Conference on Artificial Intelligence (AAAI), Washington DC, USA, 2023.

A short version of this paper also appears on International Workshop on Federated Learning: Recent Advances and New Challenges in Conjunction with NeurIPS 2022.

One-shot Neural Backdoor Erasing via Adversarial Weight Masking

Shuwen Chai# and Jinghui Chen

in Proceedings of the 36th Advances in Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA, 2022.

How Powerful is Implicit Denoising in Graph Neural Networks

Songtao Liu, Zhitao Ying, Hanze Dong, Lu Lin, Jinghui Chen and Dinghao Wu

NeurIPS 2022 Workshop on New Frontiers in Graph Learning.

LeapAttack: Hard-Label Adversarial Attack on Text via Gradient-Based Optimization

Muchao Ye, Jinghui Chen, Chenglin Miao, Ting Wang and Fenglong Ma

in Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Washington DC, USA, 2022.

Communication-Efficient Adaptive Federated Learning

Yujia Wang#, Lu Lin and Jinghui Chen

in Proceedings of the 39th International Conference on Machine Learning (ICML), Baltimore, Maryland, USA, 2022.

Learnability Lock: Authorized Learnability Control Through Adversarial Invertible Transformations

Weiqi Peng# and Jinghui Chen

in Proceedings of the 10th International Conference on Learning Representations (ICLR), Virtual, 2022.

Communication-Compressed Adaptive Gradient Method for Distributed Nonconvex Optimization

Yujia Wang#, Lu Lin and Jinghui Chen

in Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS), Virtual, 2022.

A short version of this paper also appears on International Workshop on Trustable, Verifiable and Auditable Federated Learning in Conjunction with AAAI 2022.

Efficient Robust Training via Backward Smoothing

Jinghui Chen, Yu Cheng, Zhe Gan, Quanquan Gu and Jingjing Liu

in Proceedings of the 36th AAAI Conference on Artificial Intelligence (AAAI), Vancouver, BC, Canada, 2022.

Do Wider Neural Networks Really Help Adversarial Robustness?

Boxi WuE, Jinghui ChenE, Deng Cai, Xiaofei He and Quanquan Gu

in Proceedings of the 35th Advances in Neural Information Processing Systems (NeurIPS), Virtual, 2021.

Benign Overfitting in Adversarially Robust Linear Classification

Jinghui ChenE, Yuan CaoE and Quanquan Gu

ICML 2021 Workshop on Overparameterization: Pitfalls and Opportunities.

Research

The research of our lab is focused on different aspects of machine learning (efficiency, robustness, interpretability, responsibility, trustworthiness) and their applications in computer vision, graph learning, anomaly detection, cybersecurity, recommendation systems, computational genomics, etc. Some of our current research projects are:

Trustworthy and Saftey Issues in Large Language Models
Studying the vulnerabilities inside the current LLMs and how we can improve them for better trustworthiness.
Optimization in Machine Learning
Studying the convergence of machine learning optimizers including adaptive gradient optimizers and designing new generation of optimizers for deep learning.
Poisoning/Backdoor Attacks and Defenses
Studying the effects of poisoning and backdoor attacks on deep learning models, as well as how to mitigate those threats.
Federated Machine Learning
Addressing the emerging challenges for Federated Learning in practical scenarios such as data and model heterogeneity, communication efficiency, as well as security and privacy issues.
Adversarial Robustness in Machine Learning
Evaluating, understanding, and improving adversarial robustness in deep learning as well as studying the theoretical foundations behind adversarial training and robust learning.
Robustness in Graph Nerual Networks
Improving the robustness of the current graph neural networks again graph structural/feature perturbations.

Students

Current Ph.D. Students Current Undergrad/Master/Intern Students Alumni

Teaching

Acadamic Service

Senior Program Committee Program Committee/Reviewer Journal Reviewer