David_Gerard comments on [Link] Better results by changing Bayes’ theorem - Less Wrong

3 Post author: XiXiDu 09 March 2012 07:38PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Tags: bayes

Comments (17)

Sort By: Best

You are viewing a single comment's thread. Show more comments above.

Comment author: gwern 09 March 2012 07:48:25PM 0 points [-]

Why is this interesting?

Comment author: David_Gerard 09 March 2012 08:20:53PM 2 points [-]

Erm, its relation as an example to the quote the article starts with?

Comment author: gwern 09 March 2012 08:23:56PM 9 points [-]

Any other machine learning algorithm is a potential example for the observation 'naive token-by-token Bayes can be outperformed'.

Comment author: XiXiDu 09 March 2012 08:35:54PM 8 points [-]

Any other machine learning algorithm is a potential example for the observation 'naive token-by-token Bayes can be outperformed'.

I see. I didn't know this was the case. I am sorry if this is something really banal, it was new to me. Feel free to downvote it.