2018-05-22 7:04 GMT+03:00 Nil Geisweiller <[email protected]>: > > Sounds interesting. Without details being provided that sounds to me like > meta-learning + schematization (turning the most frequent paths of a > generalized solver into a more narrow efficient program, as you did mention > during the last SingularityNET meeting). Would like to hear more about it. > > Unfortunately, only few initial steps were made in this direction, and he stopped to research this topic, at least, for now, so there are no (published) details... Initially, it was not a machine learning approach - special meta-computation techniques (based on operations of computation traces) were supposed to be used. However, I've heard he mentioned something about applying machine learning to learn specializers - this might be more similar to meta-learning + schematization. In any case, I mentioned this only as a relevant idea.
-- You received this message because you are subscribed to the Google Groups "opencog" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/opencog. To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/CABpRrhzbzAZnQD55zOmgnoehJuqik%2Byx6gcuTqknw86K0Q8XEA%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
