About me

I am Dinghuai Zhang (张鼎怀), a third year PhD student at Mila. I belong to Prof. Bengio’s group as well as Aaron’s Army (aka A.A.). I am also honored to have the opportunity to be mentored by Dr. Tian Qi Chen at FAIR labs. Previously, I was an undergraduate in School of Mathematical Sciences at Peking University, working with Prof. Zhanxing Zhu and Prof. Bin Dong.

I am always open to possible cooperation or visiting opportunities. If you are interested, please contact me by email, facebook, or wechat.

Research interests

  • Causal inference & out-of-distribution generalization (invariant prediction, adversarial robustness).
  • Uncertainty (Bayesian inference, conformal inference) & generative models.
  • Exploration with structure (RL, active learning).

Publications

Latent State Marginalization as a Low-cost Approach for Improving Exploration
Dinghuai Zhang, Aaron Courville, Yoshua Bengio, Qinqing Zheng, Amy Zhang, Ricky T. Q. Chen.
Preprint

Predictive Inference with Feature Conformal Prediction
Jiaye Teng*, Chuan Wen*, Dinghuai Zhang*, Yoshua Bengio, Yang Gao, Yang Yuan.
Preprint

Unifying Generative Models with GFlowNets
Dinghuai Zhang, Ricky T. Q. Chen, Nikolay Malkin, Yoshua Bengio.
Preprint note

Generative Flow Networks for Discrete Probabilistic Modeling [poster] [slide]
Dinghuai Zhang, Nikolay Malkin, Zhen Liu, Alexandra Volokhova, Aaron Courville, Yoshua Bengio.
39th International Conference on Machine Learning (ICML 2022)

Building Robust Ensembles via Margin Boosting [poster] [slide]
Dinghuai Zhang, Hongyang Zhang, Aaron Courville, Yoshua Bengio, Pradeep Ravikumar, Arun Sai Suggala.
39th International Conference on Machine Learning (ICML 2022)

Unifying Likelihood-free Inference with Black-box Optimization and Beyond [openreview] [slide]
Dinghuai Zhang, Jie Fu, Yoshua Bengio, Aaron Courville.
10th International Conference on Learning Representations (ICLR 2022 spotlight)

Can Subnetwork Structure be the Key to Out-of-Distribution Generalization? [poster] [slide]
Dinghuai Zhang, Kartik Ahuja, Yilun Xu, Yisen Wang, Aaron Courville.
38th International Conference on Machine Learning (ICML 2021 long talk)

Neural Approximate Sufficient Statistics for Implicit Models [openreview] [poster] [slide]
Yanzhi Chen*, Dinghuai Zhang*, Michael Gutmann, Aaron Courville, Zhanxing Zhu.
9th International Conference on Learning Representations (ICLR 2021 spotlight)

Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework [poster] [slide]
Dinghuai Zhang*, Mao Ye*, Chengyue Gong* , Zhanxing Zhu, Qiang Liu.
34th Conference on Neural Information Processing Systems (NeurIPS 2020)

Informative Dropout for Robust Representation Learning: A Shape-bias Perspective [slide] [zhihu]
Baifeng Shi*, Dinghuai Zhang*, Qi Dai, Zhanxing Zhu, Yadong Mu, Jingdong Wang.
37th International Conference on Machine Learning (ICML 2020)

You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle [poster] [slide] [zhihu]
Dinghuai Zhang*, Tianyuan Zhang*,Yiping Lu*, Zhanxing Zhu, Bin Dong.
33rd Conference on Neural Information Processing Systems (NeurIPS 2019)

( * denotes equal contribution)

Seminars

These are some of the seminars I’ve co-organized / participated in.

Miscs

  • I enjoy reading. I feel lucky to learn about wisdom from sociologists such as Georg Simmel, Norbert Elias, Max Weber, Sigmund Freud. I occasionally write some thoughts in Chinese at douban.
  • I used to be a huge fan of Daido Moriyama and Henri Cartier-Bresson (street photography pioneers).
  • I paint Chinese calligraphy well since a child.