👀 Check out @marcelhussing.bsky.social and @liv-daliberti.bsky.social's amazing new work on behavioral consistency!
❓The challenge? Retraining an RL agent might give you a completely different policy than before! This makes everything harder, as we never quite know whether we simply got unlucky 🤔
🚨 New Preprint Alert: Behavior-Consistent Deep Reinforcement Learning 🚨
TLDR: We introduce an approach that achieves behavioral similarity across independent algorithm executions in continuous state-action space deep RL.
