In various cognitive modelling publications, "One-Shot Learning", which is just learning given very few examples, is praised as being an essential part of human cognition. For example, in "One-Shot Learning with a Hierarchical Nonparametric Bayesian Model" a statistical model is used to quickly develop categories given a few examples of digits from the MNIST database.

Are models built using the NEF/SPA capable of "One-Shot Learning"? Are there any examples of this being performed in a task?

  • $\begingroup$ I think, the chance to get an answer here is very low unless there happens to be someone from the team behind the NEF on this site. Maybe it's a better idea to ask this on their mailing list ? And maybe they can post their answer here as well or you just copy it to here. $\endgroup$ – awakenting Sep 6 '16 at 14:59
  • $\begingroup$ I work in the NEF lab, which is why I usually ask the questions in the tag. This question is currently being discussed in the forum, at which point I will summarise the discussion here for future reference. $\endgroup$ – Seanny123 Sep 6 '16 at 15:01
  • $\begingroup$ Ah I see. That's convenient :) $\endgroup$ – awakenting Sep 6 '16 at 15:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.