Discovering Blind Spots in Reinforcement Learning.
Ramya RamakrishnanEce KamarDebadeepta DeyJulie A. ShahEric HorvitzPublished in: AAMAS (2018)
Keyphrases
- dynamic programming
- reinforcement learning
- state space
- markov decision processes
- optimal policy
- optimal control
- function approximation
- policy search
- learning algorithm
- robotic control
- reinforcement learning algorithms
- temporal difference
- model free
- multi agent
- machine learning
- control problems
- learning problems
- transfer learning
- image processing
- partially observable
- action space
- automatically extracting
- temporal difference learning
- real time