Agentic Reinforcement Learning
How autonomous agents learn to reason, plan, use tools, and coordinate through interaction and feedback, especially in long-horizon and multi-agent settings.
About
Hi, I'm Feixiang, an undergraduate focusing on Artificial Intelligence at SUSTech.
I am interested in how learning agents make reliable decisions: from reinforcement learning and long-horizon planning to the theory that explains why modern learning systems work.
This site is my working archive for notes, implementations, and ideas developed along the way. For more details, see my CV.
Research Interests
Learning Theory
The principles behind modern learning systems, including optimization, generalization, interpretability, and reliable decision-making.
Education
Southern University of Science and Technology
Undergraduate · Academic focus in Artificial Intelligence