User contributions for 128.148.193.121
Jump to navigation
Jump to search
5 November 2013
- 18:2918:29, 5 November 2013 diff hist +10,202 N Q-learning The Q-learning equation previously was *wrong* There are two ways to represent it and what was present was some incorrect hybrid of the two. I have corrected it. See http://webdocs.cs.ualberta.ca/~sutton/book/ebook/node65.html if you wish to validate.