FRAC-Q-Learning: A Reinforcement Learning with Boredom Avoidance Processes for Social Robots.
Akinari OnishiPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- function approximation
- real robot
- reinforcement learning algorithms
- state space
- cooperative
- model free
- social interaction
- mobile robot
- robotic control
- multi agent
- optimal policy
- machine learning
- robot control
- learning algorithm
- action selection
- agent learns
- temporal difference
- eligibility traces
- temporal difference learning
- multi robot
- collective learning
- stochastic approximation
- continuous state and action spaces
- markov decision processes
- social media
- relational reinforcement learning
- multi agent reinforcement learning
- social networks
- state action
- function approximators
- learning agent
- policy iteration
- lower bound
- partially observable
- sequential decision problems
- continuous state spaces
- multiagent learning
- continuous state
- rl algorithms
- reinforcement learning methods
- control problems
- autonomous robots
- dynamic programming