General Reinforcement Learning Algorithm: AlphaZero & AlphaGo


AlphaGo. General Reinforcement Learning Algorithm. AlphaZero vs AlphaGo Zero. AlphaZero has hard-coded rules for setting search hyperparameters. AlphaZero was trained solely via self-play. AlphaZero vs Stockfish. AlphaZero vs Stockfish results 28/0/72(W/L/D) 12 most popular human openings.

  • Technology Presentations
  • MS PowerPoint 1508 KB
  • 2018 m.
  • English
  • 16 pages (126 words)
  • University
  • Jonas
  • General Reinforcement Learning Algorithm: AlphaZero & AlphaGo
    10 - 3 votes
Citation
General Reinforcement Learning Algorithm: AlphaZero & AlphaGo. (October 23, 2018). https://documents.exchange/general-reinforcement-learning-algorithm-alphazero-alphago/ Reviewed on 06:51, February 3 2025
×