Résultats
2 articlesTrier
18:46ENDeep Learning Plus One Idea: Ilya Sutskever on Building AGIAnime : Lex FridmanIn a May 2020 conversation, Sutskever discusses self-play, simulation, bodies and consciousness, tests of intelligence, and how an AGI might be governed and aligned.
17:40ENFrom AlphaGo Zero to MuZero: How Self-Play Strips Knowledge Out of AIAnime : Lex FridmanDeepMind's David Silver explains why removing human knowledge made AlphaZero stronger and more general, and how MuZero extended the idea to worlds with unknown rules.









