LB-DESPOT: Efficient Online POMDP Planning Considering Lower Bound in Action Selection (Student Abstract).
Chenyang WuRui KongGuoyu YangXianghan KongZongzhang ZhangYang YuDong LiWulong LiuPublished in: AAAI (2021)
Keyphrases
- action selection
- lower bound
- reinforcement learning
- upper bound
- basal ganglia
- robot soccer
- online learning
- branch and bound algorithm
- objective function
- human robot
- temporal difference
- action selection mechanism
- online algorithms
- partially observable markov decision processes
- optimal solution
- planning problems
- online course
- model free
- action space
- np hard
- learning environment
- belief space
- heuristic search
- learning styles
- reinforcement learning algorithms
- partially observable
- state space
- planning under uncertainty
- multi agent
- e learning