Andrej Karpathy

Andrej Karpathy

основатель Eureka Labs
USA

Tuyên bố

  1. compute 2025-10-17
    Sharply criticizes RL as the core method: it's bad, everything else is just worse.
    «Reinforcement learning is terrible. It just so happens that everything else is much worse.»

Tranh luận có sự tham gia

Chưa có tranh luận
Tin mới