Chongxuan Li

Tenure-track Associate Professor

Chongxuan Li is an Associate Professor at the Gaoling School of Artificial Intelligence, Renmin University of China. He obtained both his Bachelor's and Ph.D. degrees from Tsinghua University. His primary research focus is on deep probabilistic learning. He has published over 40 papers in international conferences and journals in the field of machine learning, including ICML, NeurIPS, ICLR, and TPAMI. His works, such as Analytic-DPM and DPM-Solver, have been incorporated into renowned large-scale text-to-image generation models like DALL·E 2 and Stable Diffusion and were recognized as the Outstanding Paper Award at ICLR 2022 and one of the most influential papers at NeurIPS 2022 by Paper Digest respectively. Li has received accolades including the CAAI Wu Wenjun Award for Outstanding Young AI Scientist, the ACM SIGAI China Rising Star Award, the CCF Outstanding Ph.D. Thesis Award, and the first-class CAAI Wu Wenjun Award for AI. He was also selected for the China Postdoctoral Innovative Talent Support Program and Beijing Nova Program. Moreover, he will serve as an Area Chair for ICLR 2024. Homepage https://zhenxuan00.github.io/.

Enter Personal Website

Education

Tsinghua University Sep 2014 – Jul 2019 PhD in Computer Science and Technology, supervised by Bo Zhang
Tsinghua University Sep 2010 – Jul 2014 BE in Institute for Interdisciplinary Information and Sciences (Yao Class)

Work Experience

Tsinghua University Jul 2019 – Jul 2021 Postdoctoral researcher, supervised by Jun Zhu Beijing, China
University of Amsterdam Sep 2017 – Sep 2018 PhD guest researcher, supervised by Max Welling Amsterdam, Netherlands

RESEARCH INTERESTS

Deep generative models aim to learn a joint distribution on high-dimensional data. These models provide a principal and systematical way to capture the uncertainty in data and relieve the labeled-data eager problem of deep learning. Chongxuan’s research interests include: efficiently and stably learning deep generative models, applications of the models in environments with small data or limited supervision; understanding the behavior of the models in a theoretical perspective of consistency and stability.

Research Projects

National Natural Science Foundation – General Program (Nos. 62076145) Jan 2021 – Dec 2024
Study of Efficient and Convergent Learning and Inference Algorithms in Deep Generative Models Role: PI

Tsinghua-Huawei Large Granularity Long Term Corporation Project Sep 2019 – Aug 2022
Advanced Machine Learning Theory and Algorithm Role: sub-project co-leader

Chinese Postdoctoral Innovative Talent Support Program Jul 2019 – Jul 2021
Bayesian Deep Learning: Algorithms, Models and Applications Role: PI

PUBLICATIONS

2022

EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equations

Min Zhao, Fan Bao, Chongxuan Li†, Jun Zhu†

Advances in Neural Information Processing Systems (NeurIPS)

DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps

Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li and Jun Zhu

Advances in Neural Information Processing Systems (NeurIPS)

Deep reinforcement learning with credit assignment for combinatorial optimization

Dong Yan, Jiayi Weng, Shiyu Huang, Chongxuan Li, Yichi Zhou, Hang Su, Jun Zhu

Pattern Recognition

Probabilistic Neural-Symbolic Models with Inductive Posterior Constraints

Ke Su, Hang Su, Chongxuan Li, Jun Zhu, Bo Zhang

IEEE Transactions on Neural Networks and Learning Systems (TNNLS)

Fast Lossless Neural Compression with Integer-Only Discrete Flows

Siyu Wang, Jianfei Chen, Chongxuan Li, Jun Zhu and Bo Zhang

International Conference on Machine Learning (ICML)

Maximum Likelihood Training for Score-based Diffusion ODEs by High Order Denoising Score Matching

Cheng Lu, Kaiwen Zheng, Fan Bao, Chongxuan Li, Jianfei Chen and Jun Zhu

International Conference on Machine Learning (ICML)

Estimating the Optimal Covariance with Imperfect Mean in Diffusion Probabilistic Models

Fan Bao, Chongxuan Li† , Jiacheng Sun, Jun Zhu† and Bo Zhang († corresponding author)

International Conference on Machine Learning (ICML)

Memory Replay with Data Compression for Continual Learning

Liyuan Wang, Xingxing Zhang, Kuo Yang, Longhui Yu, Chongxuan Li†, Lanqing Hong, Shifeng Zhang, Zhenguo Li, Yi Zhong†, Jun Zhu† († corresponding author)

International Conference on Learning Representations (ICLR)

Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models

Fan Bao, Chongxuan Li† , Jun Zhu† and Bo Zhang († corresponding author)

International Conference on Learning Representations (ICLR) (Outstanding Paper Award, acceptence rate < 0.16%)

2021

Triple Generative Adversarial Networks

Chongxuan Li, Kun Xu, Jun Zhu, Jiashuo Liu and Bo Zhang

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI, preprint)

Stability and Generalization of Bilevel Programming in Hyperparameter Optimization

Fan Bao*, Guoqiang Wu*, Chongxuan Li*, Jun Zhu and Bo Zhang (* equal contribution)

Advances in Neural Information Processing Systems (NeurIPS)

Rethinking Univariate Losses for Multi-Label Ranking: Consistency and Generalization

Guoqiang Wu*, Chongxuan Li*, Kun Xu and Jun Zhu （* equal contribution）

Advances in Neural Information Processing Systems (NeurIPS)

ORDisCo: Effective and Efficient Usage of Incremental Unlabeled Data for Semi-supervised Continual Learning

Liyuan Wang, Kuo Yang, Chongxuan Li†, Lanqing Hong, Zhenguo Li, Jun Zhu† († corresponding author)

Conference on Computer Vision and Pattern Recognition (CVPR)

Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models

Fan Bao, Kun Xu, Chongxuan Li, Lanqing Hong, Jun Zhu and Bo Zhang

International Conference on Machine Learning (ICML)

Implicit Normalizing Flows

Cheng Lu, Jianfei Chen, Chongxuan Li, Qiuhao Wang and Jun Zhu

International Conference on Learning Representations (ICLR) (Spotlight, acceptence rate < 5.6%)

MiCE: Mixture of Contrastive Experts for Unsupervised Image Clustering

Tsung Wei Tsai, Chongxuan Li and Jun Zhu

International Conference on Learning Representations (ICLR)

2020

Understanding and Stabilizing GANs’ Training Dynamics with Control Theory

Kun Xu, Chongxuan Li, Huanshu Wei, Jun Zhu and Bo Zhang

International Conference on Machine Learning (ICML)

Learning Implicit Generative Models by Teaching Explicit Ones

Kun Xu, Chao Du, Chongxuan Li, Jun Zhu and Bo Zhang

European Conference on Machine Learning (ECML)

Efficient Learning of Generative Models via Finite-Difference Score Matching

Tianyu Pang, Kun Xu, Chongxuan Li, Yang Song, Stefano Ermon and Jun Zhu

Advances in Neural Information Processing Systems (NeurIPS)

Bi-level Score Matching for Learning Energy-based Latent Variable Models

Fan Bao*, Chongxuan Li*, Kun Xu, Hang Su, Jun Zhu and Bo Zhang (* equal contribution)

Advances in Neural Information Processing Systems (NeurIPS)

To Relieve Your Headache of Training an MRF, Take AdVIL

Chongxuan Li, Chao Du, Kun Xu, Max Welling, Jun Zhu and Bo Zhang

International Conference on Learning Representations (ICLR)

2019

Multi-objects Generation with Amortized Structural Regularization

Kun Xu, Chongxuan Li, Jun Zhu and Bo Zhang

Advances in Neural Information Processing Systems (NeurIPS)

Conditional Graphical Generative Adversarial Networks

Chongxuan Li, Jun Zhu and Bo Zhang

Journal of Software (in Chinese)

2018

Collaborative Filtering with User-Item Co-Autoregressive Models

Chao Du,Chongxuan Li, Yin Zheng, Jun Zhu and Bo Zhang

Association for the Advancement of Artificial Intelligence (AAAI)

Learning to Write Stylized Chinese Characters by Reading a Handful of Examples

Danyang Sun, Tongzheng Ren, Chongxuan Li, Jun Zhu, and Hang Su

International Joint Conferences on Artificial Intelligence (IJCAI)

Graphical Generative Adversarial Networks

Chongxuan Li, Max Welling, Jun Zhu and Bo Zhang

Advances in Neural Information Processing Systems (NeurIPS)

Max-Margin Deep Generative Models for (Semi-)Supervised Learning

Chongxuan Li, Jun Zhu and Bo Zhang

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)

2017

Population Matching Discrepancy and Applications in Deep Learning

Jianfei Chen, Chongxuan Li, Yizhong Ru and Jun Zhu

Advances in Neural Information Processing Systems (NeurIPS)

Triple Generative Adversarial Nets

Chongxuan Li, Kun Xu, Jun Zhu and Bo Zhang

Advances in Neural Information Processing Systems (NeurIPS)

2016

Towards Better Analysis of Deep Convolutional Neural Networks

Mengchen Liu, Jiaxin Shi, Zhen Li, Chongxuan Li, Jun Zhu, and Shixia Liu

IEEE VAST, TVCG track

Learning to Generate with Memory

Chongxuan Li, Jun Zhu and Bo Zhang

International Conference on Machine Learning (ICML)

2015

Max-Margin Deep Generative Models

Chongxuan Li, Jun Zhu, Tianlin Shi and Bo Zhang

Advances in Neural Information Processing Systems (NeurIPS)

Honors and Awards

ICLR Outstanding Paper Award 2022
China Computer Federation (CCF) Distinguished Ph.D. Dissertation Award
Shuimu Tsinghua Scholar Program
Chinese Postdoctoral Innovative Talent Support Program
Microsoft Research Fellowship
Runner-up prize of the vizdoom AI competition 2017 @CIG

Services

SPC: IJCAI
PC: UAI, AAAI
Reviewer: ICML, NeurIPS, ICLR, TPAMI, TNNLS

Contact

Tel：

Email：chongxuanli@ruc.edu.cn

Website：https://zhenxuan00.github.io/

Address：