Schulman Discusses Using Bad Trajectory Prefixes for RL
OpenAI co-founder John Schulman suggested that using prefixes from bad trajectories to define RL environments is a great idea, sparking a follow-up discussion with Asker Lee who shared related links.
2026-08-09 ~ 2026-08-09 · 2 related posts
- John Schulman: Using Prefixes from Misbehaving Trajectories to Define RL Environments Is a Good Idea — johnschulman2 · 2026-08-09
- Asker Lee Responds to Schulman: Shares Relevant Link — askerlee · 2026-08-09