Reinforcement Learning with Prediction-Based Rewards (via) Fascinating result: by teaching a reinforcement learning agent that plays video games to optimize for “unfamiliar states”—states where it cannot predict what will happen next—the agent does a much better job of playing some games. “... for the first time exceeds average human performance on Montezuma’s Revenge. RND achieves state-of-the-art performance, periodically finds all 24 rooms and solves the first level without using demonstrations or having access to the underlying state of the game.”
Recent articles
- Understanding GPT tokenizers - 8th June 2023
- Weeknotes: Parquet in Datasette Lite, various talks, more LLM hacking - 4th June 2023
- It's infuriatingly hard to understand how closed models train on their input - 4th June 2023
- ChatGPT should include inline tips - 30th May 2023
- Lawyer cites fake cases invented by ChatGPT, judge is not amused - 27th May 2023
- llm, ttok and strip-tags - CLI tools for working with ChatGPT and other LLMs - 18th May 2023