Nick Shea has a lovely paper arguing that 'Reward Prediction Errors are Metarepresentational' (that's the title). I am interested to find any other examples of papers relating philosophical analyses of representations, especially teleosemantics, to temporal difference learning, or reinforcement learning more generally. There seems to be surprisingly little work on this.
http://webdocs.cs.ualberta.ca/~sutton/publications.html#TD_paper
There are some good papers on the subject here on this link. I hope it helps you.
David:
There's also some more: https://www.researchgate.net/publicliterature.PublicLiterature.search.html?type=keyword&search-keyword=semantic+analysis+of+temporal+difference+learning%3F+&search-abstract=&search-journal=&search=Search
Best regards,
Debra
Check out Brown: Prediction, explanation, and the role of generative models in language processing
- Badges
- Science topic
- Similar topics
- Correction