Deep Reinforcement Learning for Pellet Eating in Agar.io.
Nil Stolt AnsóAnton Orell WieheMadalina M. DruganMarco A. WieringPublished in: ICAART (2) (2019)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- model free
- multi agent
- machine learning
- learning process
- learning algorithm
- highly efficient
- state space
- hidden markov models
- direct policy search
- objective function
- optimal policy
- learning capabilities
- belief nets
- multi agent reinforcement learning
- evolutionary learning
- continuous state
- temporal difference learning
- action space
- deep learning
- control problems
- data sets
- temporal difference
- action selection
- dynamic programming
- probabilistic model
- decision making
- evaluation function
- markov decision processes