Predicting driving behavior using inverse reinforcement learning with multiple reward functions towards environmental diversity.
Masamichi ShimosakaKentaro NishiJun-ichi SatohHirokatsu KataokaPublished in: Intelligent Vehicles Symposium (2015)
Keyphrases
- inverse reinforcement learning
- reward function
- markov decision processes
- preference elicitation
- reinforcement learning
- reinforcement learning algorithms
- state space
- optimal policy
- partially observable
- real time
- multiple agents
- markov decision process
- data mining
- state variables
- driving behavior
- transition probabilities
- dynamic systems
- np hard