EB-MC-PILCO combines iLQR trajectory optimization with the MC-PILCO learner, reducing time-to-solve on cart-pole by up to 45.9% while keeping 100% success.
Efficient reinforcement learning for robots using informative simulated priors,
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
-
Accelerating Model-Based Reinforcement Learning using Non-Linear Trajectory Optimization
EB-MC-PILCO combines iLQR trajectory optimization with the MC-PILCO learner, reducing time-to-solve on cart-pole by up to 45.9% while keeping 100% success.