【讲座】Reinforecement Learning from "Small" Data
2019.05.31 活动

You are cordially invited to the Academic Seminar to be delivered by Prof. Mengdi WANG at 9:00 am- 10:00 am on June 3rd (Monday).

 

Topic:  Reinforcement Learning from “Small” Data

Time & Date: 9:00 am - 10:00 am, Monday June 3rd

Venue:  Room 208, Cheng Dao Building

Speaker:  Prof. Mengdi WANG.  Assistant Professor, Department of Operations Research and Financial Engineering, Princeton University.

 

Abstract

Recent years have witnessed increasing empirical successes in reinforcement learning (RL). However, many theoretical questions about RL are not well understood even in the most basic setting. For example, how many observations are needed and sufficient for learning a good control policy? How to generalize from small data? In this talk, we study the statistical efficiency of reinforcement learning and show how to utilize data algorithmically in the most efficient way. For dimension reduction of high-dimensional RL, we propose a spectral state compression method for learning state embeddings from trajectorical data. We further illustrate applications of the methods to natural language processing, clinical analytics and beyond.

 

About the Speaker

 

cid:image001.png@01D51241.2CCD6490

Dr. Mengdi Wang is an assistant professor at the Department of Operations Research and Financial Engineering at Princeton University. She is also affiliated with the Department of Computer Science and Princeton’s Center for Statistics and Machine Learning. Her research focuses on data-driven stochastic optimization and applications in machine and reinforcement learning. She received her PhD in Electrical Engineering and Computer Science from Massachusetts Institute of Technology in 2013. At MIT, Mengdi was affiliated with the Laboratory for Information and Decision Systems and was advised by Dimitri P. Bertsekas. Mengdi became an assistant professor at Princeton in 2014. She received the Young Researcher Prize in Continuous Optimization of the Mathematical Optimization Society in 2016 (awarded once every three years), the Princeton SEAS Innovation Award in 2016, the NSF Career Award in 2017, the Google Faculty Award in 2017,  and the MIT Tech Review 35-Under-35 Innovation Award (China region) in 2018. She is currently serving as an associate editor for Operations Research.