Combating Reinforcement Learning's Sisyphean Curse with Intrinsic Fear.
Zachary C. LiptonJianfeng GaoLihong LiJianshu ChenLi DengPublished in: CoRR (2016)
Keyphrases
- reinforcement learning
- function approximation
- temporal difference learning
- optimal policy
- dimensionality reduction
- high dimensional
- multi agent reinforcement learning
- markov decision processes
- high dimensional data
- learning algorithm
- temporal difference
- machine learning
- real world
- database
- autonomous learning
- learning agent
- robot control
- reinforcement learning algorithms
- action selection
- high dimensionality
- case study
- state space
- geometric structure
- learning problems
- genetic algorithm
- partially observable
- dynamic programming
- real time
- learning process
- pattern recognition