Transformable Gaussian Reward Function for Socially-Aware Navigation with Deep Reinforcement Learning.
Jinyeob KimSumin KangSungwoo YangBeomjoon KimJargalbaatar YuraDonghan KimPublished in: CoRR (2024)
Keyphrases
- reward function
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- state space
- optimal policy
- multi agent
- policy search
- inverse reinforcement learning
- partially observable
- markov decision process
- multiple agents
- transition model
- maximum likelihood
- transition probabilities
- machine learning
- markov decision problems
- state action
- learning algorithm
- initially unknown
- hierarchical reinforcement learning
- model free
- temporal difference
- learning agent
- control policies
- state variables
- dynamic programming
- average reward
- action selection
- function approximation
- latent variables
- semi supervised