Jun, 2024
约束元不可知强化学习
Constrained Meta Agnostic Reinforcement Learning
Karam Daaboul, Florian Kuhm, Tim Joseph, J. Marius Zoellner
TL;DRMeta-Reinforcement Learning (Meta-RL) aims to acquire meta-knowledge for quick adaptation to diverse tasks. Our novel approach, Constraint Model Agnostic Meta Learning (C-MAML), merges meta learning with constrained optimization to enable rapid and efficient task adaptation, demonstrating effectiveness in simulated locomotion with wheeled robot tasks of varying complexity.