Advice-Guided Reinforcement Learning in a non-Markovian Environment.
Daniel NeiderJean-Raphaël GaglioneIvan GavranUfuk TopcuBo WuZhe XuPublished in: AAAI (2021)
Keyphrases
- reinforcement learning
- decision processes
- state space
- multi agent environments
- exploration strategy
- dynamic environments
- markov decision processes
- learning process
- mobile robot
- autonomous agents
- neural network
- partially observable domains
- reinforcement learning algorithms
- complex environments
- learning problems
- multi agent
- machine learning
- optimal control
- indoor environments
- function approximation
- model free
- temporal difference
- real robot
- markov decision process
- optimal policy
- real time