Imitation Is Not Enough: Robustifying Imitation with Reinforcement Learning for Challenging Driving Scenarios.
Yiren LuJustin FuGeorge TuckerXinlei PanEli BronsteinBecca RoelofsBenjamin SappBrandyn WhiteAleksandra FaustShimon WhitesonDragomir AnguelovSergey LevinePublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- function approximation
- real world
- imitation learning
- learning algorithm
- state space
- model free
- reinforcement learning algorithms
- realistic scenarios
- machine learning
- action selection
- markov decision processes
- neural network
- temporal difference
- optimal control
- computational models
- supervised learning
- data sets
- learning problems
- computational model
- evolutionary algorithm
- multi agent
- function approximators
- case study
- policy search