乌尔姆大学神经信息处理所 博士研究生
电子邮箱：zeqiang.zhang@uni-ulm.de
关于我
我目前是乌尔姆大学神经信息处理研究所的博士研究生，由 Daniel Alexander Braun 教授指导。我的研究兴趣涵盖强化学习、目标条件学习、内在动机和世界模型学习等领域。我尤其关注自主智能体如何在复杂动态的环境中，基于内部目标和好奇心进行学习与适应。此外，我也对基于智能体的建模及其在计算经济学中的应用感兴趣。
我在乌尔姆大学获得了认知系统硕士学位，该项目融合了计算机科学与心理学的跨学科内容。在此之前，我在中国的东南大学获得了电子科学与技术学士学位。
强化学习
目标条件学习
内在动机
世界模型学习
基于智能体的建模
"Simulating Labor Market Dynamics with Agent-Based Models" (Poster), Ruxin Chen and Zeqiang Zhang, 2025 Japanese Economic Association Spring Meeting, Nagoya, Japan, 2025 Link
"Adaption on the Macro Level in 2LIKE (Exam Preparation Module)" (Short Talk), Zeqiang Zhang, Workshop on Adaptive Learning in Higher Education, Ulm, Germany, 2025
"Learning Contrastively: A Novel Goal-Conditioned Supervised Learning Approach with Dual-evaluation Mechanism" (Poster), Zeqiang Zhang, Fabian Wurzberger, Gerrit Schmid, Sebastian Gottwald and Daniel Braun, FRoBio: Freiburg Robotics and Biology Conference, Freiburg, Germany, 2023 Link
"Temporal goal abstraction for goal-conditioned reinforcement learning" (Poster), Fabian Wurzberger, Zeqiang Zhang, Gerrit Schmid, Sebastian Gottwald and Daniel Braun, FRoBio: Freiburg Robotics and Biology Conference, Freiburg, Germany, 2023 Link
“From Individual Learning to Market Equilibrium: Correcting Structural and Parametric Biases in RL Simulations of Economic Models” (合作者 Ruxin Chen) 预印本论文
“Autonomous Learning from Success and Failure: Goal-Conditioned Supervised Learning with Negative Feedback” (合作者 Fabian Wurzberger, Gerrit Schmid, Sebastian Gottwald and Daniel A. Braun) 预印本论文
“Learning Robust Representations for World Models without Reward Signals” (合作者 Fabian Wurzberger, Sebastian Gottwald and Daniel A. Braun, accepted by EWRL 2025) 预印本论文（Openreview）
“Characterizing Failure Mechanism of Soft and Hard Rocks: Implication from Acoustic Emission and Machine Learning” (合作者 Zhuang Li, Nuwen Xu, Feng Gao and Biao Li)
“A Cognitive Perspective on Information Frictions in Labor Markets” (合作者 Ruxin Chen)