PPO
Technique
"The algorithm we use called PPO, you plan over every single time step. There's no hierarchy."
← All topics
1 creator
1 creator
1 creator
View all product sources →
Search all reinforcement learning matches →
Explore reinforcement learning
Related products and practices, with creator experiences across different uses.
With use and recommendations from
Proximal Policy Optimization (PPO)
Technique
Training the encoder/decoder and modulating feedback to neuron cultures playing Doom
"What we did initially was a closed loop PPO architecture. PPO is proximal policy optimization, an RL algorithm... this is primarily to keep feedback to the cells stable, and we found that the critic for PPO was particularly well aligned with how we wanted t..."
Sentry
Software
Bug tracking and error monitoring at ChatPRD; primary data source for the debugging harness
"I picked Sentry debugging — really figuring out the issue using some of our custom internal tools and then doing all the follow-up actions we do when we close bugs — as a good first harness. It has coding in it. It needed custom content and custom context."
More source quotes
Error monitoring and tracking for debugging edit tool operations
"I gave Codex access to Sentry. I gave Codex access to these edit requests. And I said slash goal Codex go through every example in Sentry, every trace in Sentry of an invalid operation on the edit tool."
About these recommendations
Products are related to this topic through catalog tags. Quotes describe the creator’s stated use; they are not necessarily recommendations for this topic.
Creator counts cover all uses and all dates. Each person counts once per product family. Paid promotions and passing mentions are excluded. Items are ordered by creator count, then name.