Andrej Karpathy explains why reinforcement learning is so inefficient — “sucking supervision through a straw.” A brutally honest …
source
Andrej Karpathy explains why reinforcement learning is so inefficient — “sucking supervision through a straw.” A brutally honest …
source