CO-PILOT: COllaborative Planning and reInforcement Learning On sub-Task curriculum.
Shuang AoTianyi ZhouGuodong LongQinghua LuLiming ZhuJing JiangPublished in: NeurIPS (2021)
Keyphrases
- reinforcement learning
- action selection
- goal oriented
- function approximation
- learning algorithm
- partially observable
- state space
- macro actions
- collaborative learning
- planning process
- cooperative learning
- optimal policy
- heuristic search
- markov decision processes
- peer assessment
- complex domains
- machine learning
- dynamic programming
- multi agent
- learning classifier systems
- computer supported collaborative learning
- high school
- temporal difference
- knowledge sharing
- domain independent
- partially observable markov decision processes
- control system
- reinforcement learning methods
- neural network