Stuck on Repeat: When Reinforcement Learning Fails to Notice the Rules Changed
Opening — Why this matters now Reinforcement learning has a credibility problem. Models ace their benchmarks, plots look reassuringly smooth, and yet the moment the environment changes in a subtle but meaningful way, performance falls off a cliff. This is usually dismissed as “out-of-distribution behavior” — a polite euphemism for we don’t actually know what our agent learned. ...