On Mon, Nov 24, 2014 at 2:15 PM, Jim Bromer <[email protected]> wrote: An algorithm that could 'predict' bits in arbitrary sequences which it had never been exposed to can hardly be called a 'learning algorithm'.
On Mon, Nov 24, 2014 at 8:24 PM, Matt Mahoney via AGI <[email protected]> wrote: Yes it is. Are you familiar with AIXI? -- Matt Mahoney, [email protected] I remember you and other people talking about it. I find it interesting that you did not dodge my fundamental criticism. So that is probably where our disagreement is situated. Before you go on let me say one thing. A human calculator does not have to be exposed to every possible resultant that he could calculate in order to understand the method. But that method is not a 'learning algorithm'. It might be called a learned algorithm or something. Jim Bromer On Mon, Nov 24, 2014 at 8:24 PM, Matt Mahoney via AGI <[email protected]> wrote: > On Mon, Nov 24, 2014 at 2:15 PM, Jim Bromer <[email protected]> wrote: >> An algorithm that could 'predict' bits in arbitrary sequences which it >> had never been exposed to can hardly be called a 'learning algorithm'. > > Yes it is. Are you familiar with AIXI? > > -- > -- Matt Mahoney, [email protected] > > > ------------------------------------------- > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/24379807-653794b5 > Modify Your Subscription: https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
