Career Profile

I got my Bachelor degree and Master degree in Harbin Institute of Technology. My bachelor thesis is about model-based MARL and supervisored by Prof. Dibangoye. My research interests lie in Reinforcement Learning, particularly in Multi-Agent RL.

I am currently working at Tencent as a reinforcement learning engineer.

News

Aug 2022
The paper named Correcting Biased Value Estimation in Mixing Value-Based Multi-Agent Reinforcement Learning by Multiple Choice Learning is accepted by Engineering Applications of Artificial Intelligence (IF:7.802/Q1).
July 2022
Joined Tencent as full-time reinforcement learning engineer.
July 2022
The paper named Multi-level credit assignment for cooperative multi-agent reinforcement learning is accepted by Applied Sciences (IF:2.838/Q2).
December 2020
Joined Netease Fuxi AI lab as intern.
July 2020
Joined a great company Parametrix.ai that focus on applying AI in games.
June 2020
Optimally Solving Two-Agent Decentralized POMDPs Under One-Sided Information Sharing accepted by ICML 2020.
February 2020
Submitted a paper to ICML 2020.
September 2019
Joined CITI-Lab in INRIA and INSA de Lyon and studied on MARL under the supervision of Prof. Dibangoye.
April 2019
Paper accepted at IIHMSP in Jilin China.

Publications

Liu Bing, Xie Yuxuan, Feng Lei, Fu Ping
Engineering Applications of Artificial Intelligence, 2022
Feng Lei, Xie Yuxuan, Liu Bing, Wang Shuyan
Applied Sciences, 2022
Xie Yuxuan, Jilles S. Dibangoye, Olivier Buffet
ICML, 2020
Xie Yuxuan, Liu Bing, Feng Lei, Li Xipeng, Zou Danyin.
IIHMSP/FITAT, 2019

Education

Master

2020 - 2022
Harbin Institute of Technology

Exchange student funded by CSC

2019 - 2020
INSA de Lyon

Exchange student funded by HIT

2018 - 2018
Peking University

Bachelor Degree (Ranking:1/110, GPA:93.5/100)

2016 - 2020
Harbin Institute of Technology

Experiences

RL Internship

2020.12-2021.3
Netease Fuxi LAB

RL Internship

2020.7-2020.9
Parametrix.ai

Research Assistant

2019.9-2020.6
Chroma, CITI LAB, INSA-Lyon & INRIA
  • Proposed belief occupancy state as a summary to recast Dec-POMDPs under one-sideness sharing to boMDP which is MDP actually.
  • Implemented belief occupancy state Heuristic search and value iteration algorithm to solve boMDP.
  • Applied linear programming and tabular method to improve the scalability.

Lead Developer

2018-2019
Auto Test and Control Lab
  • Quantized floating point data of DL Networks into 16 or 8 bits on Caffe.
  • Applied KL Divergence to decrease the loss caused by quantization of 8 bits to just 1.5 for MobileNet-SSD.
  • Verified the Quantization Scheme for 16 bits on FPGA.

Developer

2018-2019
Auto Test and Control Lab
  • Designed and trained DL Network for diagnosis of pneumonia on Caffe and implemented it on Zynq.
  • Won 2nd Place in the 16th Challenge Cup and Silver Award in the 9th Zuguang Cup.

Developer

2018-2018
Wireless Charging Lab
  • Manufactured and debugged control module for a wireless charging system.