An ML Agent using the Policy Gradient Method to win a SoccerTwos Game.
Victor Ulisses PugliesePublished in: ICEIS (1) (2022)
Keyphrases
- gradient method
- policy gradient
- actor critic
- action selection
- convergence rate
- agent learns
- multi agent
- multi agent systems
- maximum likelihood
- reinforcement learning
- convex formulation
- negative matrix factorization
- pricing mechanism
- state action
- multiple agents
- optimization methods
- stochastic games
- step size
- reward function
- optimal policy
- single agent
- markov decision process
- natural gradient learning
- machine learning
- policy iteration
- sparse representation
- optimization algorithm
- image classification
- signal processing
- text mining