Automatic collective motion tuning using actor-critic deep reinforcement learning.
Shadi AbpeikarKathryn KasmarikMatthew GarrattRobert A. HunjetMd Mohiuddin KhanHuanneng QiuPublished in: Swarm Evol. Comput. (2022)
Keyphrases
- actor critic
- reinforcement learning
- temporal difference
- policy gradient
- approximate dynamic programming
- optimal control
- reinforcement learning algorithms
- gradient method
- neuro fuzzy
- function approximation
- policy iteration
- state space
- image sequences
- policy gradient methods
- model free
- learning algorithm
- natural actor critic
- markov decision processes
- optical flow
- control problems
- average reward
- optimal policy
- partially observable
- markov decision process
- transfer learning
- convergence rate
- monte carlo
- reinforcement learning methods
- rl algorithms
- dynamic programming
- fuzzy rules
- fixed point
- motion planning