An online follow-the-leader world model updated by analytic least squares, combined with CEM planning, solves sequential robotic tasks without forgetting on a new unified-dynamics benchmark.
Gradient based sample selection for online continual learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Continual Reinforcement Learning by Planning with Online World Models
An online follow-the-leader world model updated by analytic least squares, combined with CEM planning, solves sequential robotic tasks without forgetting on a new unified-dynamics benchmark.