Hi, I'm Yitao Liang

A CS Ph.D student at UCLA, focusing on machine learning.

More About Me

yliang@cs.ucla.edu Google Scholar Github LinkedIn

About Me

Mastery itself is the prize of the venture.

I was born and raised in Nanjing, a historical yet dynamic city along Yangtze river in China. I am very fortunate to be a member of the Statistical and Relational Artificial Intelligence (StarAI) Lab, led by Guy Van den Broeck. My research passion is on "learning and making decision under structured environments with uncertainty". To this end, I am dedicated to making contributions on two frontiers. The first one emphasizes on how to make good decisions; we develop and leverage cutting-edge reinforcement learning (RL) techniques to achieve this. The other frontier is focused on how to model the structure, domain knowledge and uncertainty embedded in the environment; we invent novel probabilistic graphical models and differentiable loss functions for that.

Before joining UCLA, I received my Bachelor degree from Franklin & Marshall College, working under the advisement of Erin Talvitie.

I am on the job market this year. If you are interested, please check out my research statement .

Working Experience

UCLA

Research Assistant, September 2016 - Present

Facebook AI

Research Intern, June - September 2018

Amazon

Software Engineer, May - September 2016

Papers

Publications

On Effective Parallelization of Monte Carlo Tree Search

Anji Liu, Yitao Liang, Ji Liu, Guy Van den Broeck, Jianshu Chen
Preprint, 2021
More details: Paper

Juice: A Julia Package for Logic and Probabilistic Circuits

Meihua Dang, Pasha Khosravi, Yitao Liang, Antonio Vergari, Guy Van den Broeck
Proceedings of the 35th AAAI Conference on Artificial Intelligence (Demo Track), 2020
More details: Paper, Bibtex

SAM: Squeeze-and-Mimic Networks for Conditional Visual Driving Policy Learning

Albert Zhao, Tong He, Yitao Liang, Haibin Huang, Guy Van den Broeck, Stefano Soatto
Conference on Robot Learning, 2020
More details: Paper, Bibtex

Handling Missing Data in Decision Trees: A Probabilistic Approach

Pasha Khosravi, Antonio Vergari, Yoojung Choi, Yitao Liang, Guy Van den Broeck
The Art of Learning with Missing Values Workshop at ICML (Artemiss), 2020
More details: Paper, Bibtex

Off-Policy Deep Reinforcement Learning with Analogous Disentangled Exploration

Anji Liu, Yitao Liang, Guy Van den Broeck
Proceedings of AAMAS, 2020
More details: Paper, Bibtex

On Tractable Computation of Expected Predictions

Pasha Khosravi, YooJung Choi, Yitao Liang, Antonio Vergari, Guy Van den Broeck
Proceedings of NeurIPS, 2019
More details: Paper, Code, Slides, Bibtex

What to Expect of Classifiers? Reasoning about Logistic Regression with Missing Features

Pasha Khosravi, Yitao Liang, YooJung Choi, Guy Van den Broeck
Proceedings of IJCAI, 2019
More details: Paper, Code, Slides, Bibtex

Learning Logistic Circuits

Yitao Liang, Guy Van den Broeck
Proceedings of AAAI, 2019
Oral Full Presentation (5.9%)
More details: Paper, Code, Slides, Bibtex

Horizon: Facebook’s Open Source Applied Reinforcement Learning Platform

Jason Gauci, Edoardo Conti, Yitao Liang, Kittipat Virochsiri, Yuchen He, Zachary Kaden, Vivek Narayanan, Xiaohui Ye
Proceedings of RL for Real Life Workshop in ICML, 2019
Best Paper (6.6%)
More details: Post, Paper, Code, Bibtex

Inference Aided Reinforcement Learning for Incentive Mechanism Design in Crowdsourcing

Zehong Hu, Yitao Liang, Jie Zhang, Zhao Li, Yang Liu
Proceedings of NeurIPS, 2018
More details: Paper, Code, Bibtex

A Semantic Loss Function for Deep Learning with Symbolic Knowledge

Jingyi Xu, Zilu Zhang, Tal Friedman, Yitao Liang, Guy Van den Broeck
Proceedings of ICML, 2018
More details: Paper, Code, Bibtex

A Semantic Loss Function for Deep Learning under Weak Supervision

Jingyi Xu, Zilu Zhang, Tal Friedman, Yitao Liang, Guy Van den Broeck
Proceedings of LLD Workshop in NIPS, 2017
Best Paper Runner-up (6.5%)

Towards Compact Interpretable Models: Shrinking of Learned Probabilistic Sentential Decision Diagrams

Yitao Liang, Guy Van den Broeck
Proceedings XAI Workshop in IJCAI, 2017
More details: Paper

Learning the Structure of Probabilistic Sentential Decision Diagrams

Yitao Liang, Jessa Bekker, Guy Van den Broeck
Proceedings of UAI, 2017
Oral Full Presentation (8.4%)
More details: Paper, Code, Slides, Bibtex

State of the Art Control of Atari Games Using Shallow Reinforcement Learning

Yitao Liang, Marlos C. Machado, Erik Talvitie, Michael Bowling
Published in AAMAS, 2016
Best Paper Nomination (0.7%)
More details: Paper, Code, Bibtex