This early morning I devoted a while to test an idea I have had in mind for years. In fact, I’ve told Edanna several times:

Why don’t we use Markov’s chains to implement several learning parts of an AI software? I’ve always thought human-machine interaction (a relatively new field which offers big promises) can borrow much of Markov’s ideas for its application.

When we learn our mother tongue, in some way, or neurons learn to expect that after a words sequence only a reduced sets of words can follow. And, in some manner, this is what I’m going o do next.

Read the rest of this entry »

Posted by Boriel as Computer Science, Artificial Intelligence, Mathematics at 12.06 pm

Rate This Post:

(1 votes, average: 5 out of 5)