Hai Zhang
♂
PERSONAL STATEMENT
A dedicated, practicing, and capable researcher in reinforcement learning (RL), specifically in model-based RL,
offline RL and context-based meta RL, with absorbing more than 100 papers in related fields (see the details in
zhihu). A confident and cohesive member in collaborative work, brilliant at communication and organization.
A pragmatic and skillful back-end development engineer before diving into RL, with the mastering of C++,
PYTHON, Golang, and some common techniques.
Education
TONGJI UNIVERSITY, SHANGHAI, CHINA
2022.09 – up to now
2018.09 – 2022.06
MASTER(without examination), COMPUTER SCIENCE AND TECHNOLOGY
with GPA of more than 85
TONGJI UNIVERSITY, SHANGHAI, CHINA
BACHELOR, COMPUTER SCIENCE AND TECHNOLOGY
with GPA of more than 92, general ranking of 5.12%
Research experience
*
means co-first author, † means corresponding author
ZHEJIANG LAB HANGZHOU, ZHEJIANG, CHINA
RESEARCH INTERN supervised by Lanqing Li
2023.07 – 2024.01
•
Towards an Information Theoretic Framework of Context-Based Offline Meta-Reinforcement Learn-
ing. Lanqing Li*, Hai Zhang*, Xinyu Zhang, Shatong Zhu, Junqiao Zhao†, and Pheng-Ann Heng. Under
review in ICML 2024
TIEV LAB SHANGHAI, CHINA
2022.09 – up to now
•
How to Fine-tune the Model: Unified Model Shift and Model Bias Policy Optimization. Hai Zhang,
Hang Yu, Junqiao Zhao†, Di Zhang, Chang Huang, Hongtu Zhou, Xiao Zhang and Chen Ye. In NeurIPS
023
2
•
Safe Reinforcement Learning with Dead-Ends Avoidance and Recovery. Xiao Zhang, Hai Zhang,
Hongtu Zhou, Chang Huang, Di Zhang, Chen Ye†, Junqiao Zhao†. In IEEE Robotics and Automation
Letters 2023 (Present in ICRA 2024)
Projects
Distributed Complete Vehicle Cloudization
2022.05 – 2023.01
•
•
•
Invention Patent (Submitted, Patent Number: 202310899331.2)
Affiliated with the National Key R&D Program
Core Technique: distributed framework architecture, k8s deployment
NIO, SHANGHAI, CHINA
BACKEND DEVELOPMENT ENGINEER
2021.10-2022.03
•
LIFELONG CYCLE MANAGE SYSTEM (MongoDB replica set, MongoDB atomic operation, Nginx re-
verse proxy, docker deployment, Redis distributed lock)
•
•
•
EDGE-SIDE DATA UPLOAD AND DISPLAY (Protobuf, web-socket, TTL index, split-table storage)
DATA SYNCHRONIZATION SERVICE (Kafka consumer group, full and increment synchronization)
5T MAGNITUDE DATA MIGRATION AND BACKUP (MongoDB shared set, shared key setting, index
setting, MongoDB cluster construction)