Yifan Wu

Email: ywu12backup@gmail.com

github

I am currently working on automated trading systems.

I completed my PhD in the Machine Learning Department at Carnegie Mellon University in 2021. My advisor is Zachary Lipton.

I finished my Master program in the Department of Computing Science at the University of Alberta working with Csaba Szepesvári and András György in 2016. I received my Bachelor's degree in Computer Science from Shanghai Jiao Tong University in 2013.

Internships I did in the past: 08/2017-11/2017 at Google DeepMind, London; 05/2018-08/2018, 05/2019-08/2019, 05/2020-08/2020 at Google Brain, Mountain View.

Research

My PhD research was focusing on fundamental problems in pushing machine learning into practical use. I have worked on (i) reinforcement learning and decision making, (ii) prediction under distribution shift.

Papers

Mixture Proportion Estimation and PU Learning: A Modern Approach. pdf
Saurabh Garg, Yifan Wu, Alex Smola, Sivaraman Balakrishnan and Zachary Lipton.
In NeurIPS 2021.

On the Optimality of Batch Policy Optimization Algorithms. pdf
Chenjun Xiao*, Yifan Wu*, Tor Lattimore, Bo Dai, Jincheng Mei, Lihong Li, Csaba Szepesvari, Dale Schuurmans. (*equal contribution)
In ICML 2021.

Instabilities of Offline RL with Pre-Trained Neural Representation. pdf
Ruosong Wang, Yifan Wu, Ruslan Salakhutdinov, Sham M. Kakade.
In ICML 2021.

Learning Local Advantage Functions for Generalizable Graph Optimizations. pdf
Yifan Wu, Yanqi Zhou, Phitchaya Mangpo Phothilimthana, Hanxiao Liu, Sudip Roy, Azalia Mirhoseini.
In NeurIPS 2020 Machine Learning for Systems Workshop.

A Unified View of Label Shift Estimation. pdf
Saurabh Garg, Yifan Wu, Sivaraman Balakrishnan, Zachary Lipton.
In NeurIPS 2020.

Behavior Regularized Offline Reinforcement Learning. pdf code
Yifan Wu, George Tucker, Ofir Nachum.
In NeurIPS 2019 Deep Reinforcement Learning Workshop.

Learning to Combat Compounding-Error in Model-Based Reinforcement Learning. pdf
Chenjun Xiao*, Yifan Wu*, Chen Ma, Dale Schuurmans, Martin Mueller. (*equal contribution)
In NeurIPS 2019 Deep Reinforcement Learning Workshop.

Game Design for Eliciting Distinguishable Behavior. pdf
Fan Yang, Leqi Liu, Yifan Wu, Zachary Lipton, Pradeep Ravikumar, Tom Mitchell, William Cohen.
In NeurIPS 2019.

Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment. pdf code
Yifan Wu, Ezra Winston, Divyansh Kaushik, Zachary Lipton.
In ICML 2019.

The Laplacian in RL: Learning Representations with Efficient Approximations. pdf code
Yifan Wu, George Tucker and Ofir Nachum.
In ICLR 2019.

Towards Understanding the Generalization Bias of Two Layer Convolutional Linear Classifiers with Gradient Descent. pdf
Yifan Wu, Barnabás Póczos and Aarti Singh.
In AISTATS 2019.

Importance Reweighting Using Adversarial-Collaborative Training. pdf
Yifan Wu, Tianshu Ren and Lidan Mu.
In NIPS 2016 Workshop on Adversarial Training.

Conservative Bandits. pdf
Yifan Wu, Roshan Shariff, Tor Lattimore and Csaba Szepesvári.
In ICML 2016.

Online Learning with Gaussian Payoffs and Side Observations. pdf
Yifan Wu, András György and Csaba Szepesvári.
In NIPS 2015.

On Identifying Good Options under Combinatorially Structured Feedback in Finite Noisy Environments. pdf
Yifan Wu, András György and Csaba Szepesvári.
In ICML 2015.

Dynamic Monitoring of Optimal Locations in Road Network Databases. pdf
Bin Yao, Xiaokui Xiao, Feifei Li, Yifan Wu
In The VLDB Journal (2014) 23:697–720.

Activities

Reviewer, NIPS 2015, AISTATS 2016, JMLR, ICML 2016, ICML 2017, NIPS 2017, AAAI 2018, NeurIPS 2019, NeurIPS 2020, ICML2021.