Googleã®DeepMindãŒå›²ç¢ã®ãƒ«ãƒ¼ãƒ«ã‚’自らå¦ç¿’ã™ã‚‹ã“ã¨ã§AlphaZeroを凌駕ã™ã‚‹å¼·åŒ–å¦ç¿’手法「Muzeroã€ã‚’æ案。3ã¤ã®è¦ç‚¹ ✔ï¸ãã®ï¼‘ DeepMindã‹ã‚‰AlphaZeroã®é€²åŒ–版「MuZeroã€ãŒç™»å ´ ✔ï¸ãã®ï¼’ モデルベース強化å¦ç¿’ã«ã‚ˆã‚Šãƒ«ãƒ¼ãƒ«ã‚’与ãˆãªãã¦ã‚‚AlphaZeroã«å‹åˆ© ✔ï¸ãã®ï¼“ 囲ç¢ãƒ»ãƒã‚§ã‚¹ãƒ»å°†æ£‹ã«åŠ ãˆã¦Atariã§ã‚‚åŒä¸€ãƒ¢ãƒ‡ãƒ«ã§æœ€é«˜æ€§èƒ½ã‚’é”æˆ ç¶šãã‚’èªã‚€ã«ã¯ (3155æ–‡å—ç”»åƒ6æžš) AI-SCHOLAR㫠登録ã„ãŸã ãå¿…è¦ãŒã‚ã‚Šã¾ã™ã€‚ 1分ã§ç„¡æ–™ã§ç°¡å˜ç™»éŒ²ã™ã‚‹ ã¾ãŸã¯ ãƒã‚°ã‚¤ãƒ³
{{#tags}}- {{label}}
{{/tags}}