Flappy-Bird-AI
A small RL project that trains a PPO policy to play Flappy Bird — a sandbox for experimenting with reward shaping, observation design, and stable policy-gradient training on a control problem with sparse, repetitive feedback.
Yuanpeng Li
PhD student in Statistics at UC Irvine. Multi-task RL and LLM-to-MoE distillation.
- Irvine, CA
- University of California, Irvine
- Google Scholar
- Github