
Xiaotian Hao
College of Intelligence and Computing, Tianjin University
reinforcement learning
transfer learning
curriculum learning
bioinformatics
applications
mcts
multiagent learning
reinforcement learning algorithms
natural sciences
alphazero
muzero
multiagent gumbel alphazero
multiagent gumbel muzero
combinatorial action spaces
3
presentations
79
number of views
SHORT BIO
Hao Xiaotian is a fourth-year doctoral student in the College of Intelligence and Computing at Tianjin University, supervised by Professor Hao Jianye. His research focuses on multi-agent reinforcement learning and combinatorial optimization. He has published six first-authored papers at top international conferences such as ICML, NeurIPS, and ICLR. Additionally, he serves as a reviewer for multiple conferences. He has interned at Alibaba's Core Advertising Algorithm Team, Huawei's Enterprise Intelligence Team, and Huawei's Decision Reasoning Research Team, gaining extensive practical experience. He proposed a Simplex algorithm based on binary trees, which achieved a more than twofold acceleration in linear programming solving time when applied to Huawei's core production scheduling problem with tens of millions of variables. He was previously recognized as an outstanding intern at Huawei's Noah's Ark Laboratory and received the National Scholarship for Graduate Students.
Presentations

PORTAL: Automatic Curricula Generation for Multiagent Reinforcement Learning | VIDEO
Jizhou Wu and 6 other authors

Designing Biological Sequences without Prior Knowledge Using Evolutionary Reinforcement Learning
Xi Zeng and 6 other authors

Multiagent Gumbel MuZero: Efficient Planning in Combinatorial Action Spaces
Xiaotian Hao and 5 other authors