Game-Inspired Reinforcement Research
Agents learn by iterating through curated game scenarios.Motivation
Reinforcement learning mirrors behavioral psychology: agents learn by trial, error, and delayed rewards. We built game-style simulation environments to analyze how curriculum design, reward shaping, and exploration strategies affect learning speed and transferability.
Contributions
- Custom gym environments that emulate human problem-solving tasks
- Experiments with curiosity-driven and mistake-based reward signals
- Insights that informed later neuromorphic RL work, where biological realism and safety are crucial
Description
Reinforcement learning is an area of machine learning and is inspired by behavioral psychology. This algorithm mimics the process which the humans learn something from mistake. As this algorithm is similar with the way the humans think, it is lekely to be used in area of strong AI.
Cheolsoo Park
Professor
His research interests include machine learning, adaptive signal processing, computational neuroscience, and wearable technology.