
Position: Deployed Reinforcement Learning should be Continual
ICML 2026
Paper
Poster
Presentation
Many deployments are continual RL problems approximated as non-continual. We term these ‘measurable deployments’.

Many deployments are continual RL problems approximated as non-continual. We term these ‘measurable deployments’.

Background planning with option models allows for faster credit assignment and adaptation.
