John Schulman
A researcher at OpenAI and the originator of the Proximal Policy Optimization (PPO) algorithm, recognized as a prominent contributor to the field of reinforcement learning.
Timeline (last 90 days)
John Schulman explained that when doing AI research one usually does not directly optimize the objective one ultimately cares about, because obtaining feedback is extremely expensive and infrequent, and labs currently rely heavily on human intuition.
John Schulman is the Chief Scientist at Thinking Machines, and also a former OpenAI co-founder and core contributor to RLHF.
Everything that can be learned through RL is essentially a small number of bits that can be easily distilled and replicated, which is the primary force countering the centralization of model providers.
All Facts (3)
John Schulman explained that when doing AI research one usually does not directly optimize the objective one ultimately cares about, because obtaining feedback is extremely expensive and infrequent, and labs currently rely heavily on human intuition.
50%UnverifiedJohn Schulman is the Chief Scientist at Thinking Machines, and also a former OpenAI co-founder and core contributor to RLHF.
50%UnverifiedEverything that can be learned through RL is essentially a small number of bits that can be easily distilled and replicated, which is the primary force countering the centralization of model providers.
50%