Ruiyi Zhang @ Duke CS

Modified: 2014/08/25 22:19 by admin - Uncategorized
Ruiyi Zhang
Chinese name
Ph.D. Candidate
Department of Computer Science
Duke University

Laboratory: D344, LSRC and iiD, Gross Hall, West Campus of Duke University
Email: ryzhang #


About Me

I am currently a final-year Ph.D. candidate at the Department of Computer Science, Duke University.

Before that, I received B.Sc. degree in the Department of Computer Science from Nanjing University in 2016. I was a member of LAMDA Group, led by Dr. Zhi-Hua Zhou. I worked as a research intern at Google Brain (Mountain View, Summer 2019), Samsung Research America (Mountain View, Spring 2019 & 2020), Adobe Research (San Jose, Summer 2018), Alibaba AntAI (Hangzhou, Summer 2016).

Supervisor: Lawrence Carin.
Committee Members: Ron Parr, Rong Ge and Yiran Chen.

My research interests include machine learning, reinforcement learning (RL) and natural language processing (NLP), especially the intersection of them. More specifically, I focused on the subfields:
  • Learning from Sequences: Text Generation and Intelligent Interactive Systems;
  • Uncertainty Estimation in Deep RL;
  • Scalable Bayesian sampling and inference.



  • Sep. 2020: Three papers have been accepted to EMNLP 2020.
  • May 2020: Two papers (one with Google & DeepMind) have been accepted to ICML 2020.
  • Apr. 2020: One paper has been accepted to ACL 2020.
  • Jan. 2020: Two papers have been accepted to AISTATS 2020.
  • Dec. 2019: Two papers (one with Google Brain) have been accepted to ICLR 2020.
  • Nov. 2019: One paper has been accepted to AAAI 2020.
  • Sep. 2019: Two papers (one with Samsung Research) have been accepted to NeurIPS 2019.
  • Jun. 2019: I was lucky to be invited to present our recent works in Workshop on Stein's Method @ ICML2019 [Slides]
  • May 2019: I passed my Preliminary Exam, and became a Ph.D. Candidate!
  • Apr. 2019: Two papers have been accepted to ICML 2019.


Selected Publications [Full List]

Learning from Sequences

  • Improving Adversarial Text Generation by Modeling the Distant Future [Paper]
    Ruiyi Zhang, Changyou Chen, Zhe Gan, Wenlin Wang, Zheng Wen, Lawrence Carin

    Annual Conference of the Association for Computational Linguistics (ACL), 2020. 
    Abridged in ICML 2019, Real-World Sequential Decision-Making Workshop
  • Text-based Interactive Recommendation via Constraint Augumented Reinforcement Learning [Paper]
    Ruiyi Zhang, Tong Yu, Yilin Shen, Hongxia Jin, Changyou Chen, Lawrence Carin
    Neural Information Processing Systems (NeurIPS), 2019. 
  • Semantic Matching for Sequence-to-sequence Learning [Paper]
    Ruiyi Zhang, Changyou Chen, Xinyuan Zhang, Ke Bai, Lawrence Carin
    The Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020. 
  • Nested-Wasserstein Self-Imitation Learning for Sequence Generation [Paper]
    Ruiyi Zhang, Changyou Chen, Zhe Gan, Zheng Wen, Wenlin Wang, Lawrence Carin
    Artificial Intelligence and Statistics (AISTATS), 2020. 
    Abridged in NeurIPS 2019, Deep Reinfoecement Learning Workshop
  • Figure Captioning with Relation Maps for Reasoning [Paper]
    Charles Chen, Ruiyi Zhang, Eunyee Koh, Sungchul Kim, Scott Cohen, Ryan Rossi
    Winter Conference on Applications of Computer Vision (WACV), 2020. 
  • Topic-Guided Variational Auto-Encoders for Text Generation (Oral) [Paper]
    Wenlin Wang, Zhe Gan, Hongteng Xu, Ruiyi Zhang, Dinghan Shen, Guoyin Wang, Changyou Chen, Lawrence Carin
    North American Chapter of the Association for Computational Linguistics (NAACL), 2019. 
  • Improving Sequence-to-Sequence Learning via Optimal Transport [Paper] [Code]
    Liqun Chen, Yizhe Zhang, Ruiyi Zhang, Chenyang Tao, Zhe Gan, Bai Li, Dinghan Shen, Changyou Chen, Lawrence Carin
    International Conference for Learning Representations (ICLR), 2019. 

Uncertainty Estimation in RL

  • GenDICE: Generalized Offline Estimation of Stationary Values (Oral) [Paper]
    Ruiyi Zhang*, Bo Dai*, Lihong Li, Dale Schuurmans
    International Conference on Learning Representations (ICLR), 2020. 
  • Influence Diagram Bandits: Variational Thompson Sampling for Structured Bandit Problems [Paper]
    Tong Yu, Branislav Kveton, Zheng Wen, Ruiyi Zhang, Ole J Mengshoel
    International Conference on Machine Learning (ICML), 2020. 
  • Bayesian Meta Sampling for Fast Uncertainty Adaptation [Paper]
    Zhenyi Wang, Yang Zhao, Ping Yu, Ruiyi Zhang, Changyou Chen
    International Conference on Learning Representations (ICLR), 2020. 
  • Scalable Thompson Sampling via Optimal Transport [Paper] [Poster] [Code]
    Ruiyi Zhang, Zheng Wen, Changyou Chen, Chen Fang, Tong Yu, Lawrence Carin
    Artificial Intelligence and Statistics (AISTATS), 2019. 
    Abridged in NeurIPS 2018, Workshop on Infer to Control
  • Policy Optimization as Wasswerstein Gradient Flows [Paper] [Poster] [Appendix] [Demo]
    Ruiyi Zhang, Changyou Chen, Chunyuan Li, Lawrence Carin
    International Conference on Machine Learning (ICML), 2018. 
    Abridged in NIPS 2017, Workshop on Deep Reinforcement Learning
  • Learning Structural Weight Uncertainty for Sequential Decision-Making [Paper] [Poster] [Code]
    Ruiyi Zhang, Chunyuan Li, Changyou Chen, Lawrence Carin
    Artificial Intelligence and Statistics (AISTATS), 2018. 


Professional Sevices


Awards & Honors

  • Outstanding Graduate Student of Nanjing University, GPA 1st/143 in CS, 2016
  • CCF (China Computer Federation) Outstanding Undergraduate Award 2015
  • Microsoft Young Fellowship by Microsoft Research Aisa 2015
  • Pacemaker of Outstanding Student of Nanjing University (30/12000+) 2014
  • National Undergraduate Scholarship for consecutive three years. 2013


Outside of research, I enjoy reading, traveling and cooking. I am an amateur photographer and put some photos on 500px .

D344 LSRC, Box 90129, Duke University, Durham, North Carolina 27708-0129

The end