Ai Reinforcement Learning

Research Frontiers in Reward Model Research

Doctoral work examines learned proxies standing in for true objectives. Proxy models are imperfect and can be exploited by strong optimisation.

All AI Reinforcement Learning PhD categories