Dota 2 with Large Scale Deep Reinforcement Learning.
Christopher BernerGreg BrockmanBrooke ChanVicki CheungPrzemyslaw DebiakChristy DennisonDavid FarhiQuirin FischerShariq HashmeChristopher HesseRafal JózefowiczScott GrayCatherine OlssonJakub PachockiMichael PetrovHenrique Pondé de Oliveira PintoJonathan RaimanTim SalimansJeremy SchlatterJonas SchneiderSzymon SidorIlya SutskeverJie TangFilip WolskiSusan ZhangPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- small scale
- markov decision processes
- multi agent
- learning algorithm
- real world
- state space
- real life
- learning process
- artificial intelligence
- machine learning
- real time
- reinforcement learning algorithms
- search space
- similarity measure
- supervised learning
- learning classifier systems
- model free
- action selection
- markov decision process
- deep learning
- temporal difference learning