Since a decade ago when new life was breathed in to neural networks in the form of deep learning a plethora of different architectures have come about. Is there a reference that gives compendium of different architectures, history and target goals for each architecture (it would also be nice to know what theory is used to judge when to pick one over another)?

  • $\begingroup$ There are two kinds : those that work on a real problem for which there is no theoretical model, and those that work on data generated by some theoretical model (for example ARMA) but not on real life problems. In the latter case NN are mostly suboptimal compared to machine learning,probabilistic and signal processing methods. $\endgroup$ – reuns Aug 17 at 5:02
  • $\begingroup$ @reuns Could you provide reference justifications of 'your' way of seeing things? $\endgroup$ – T.... Aug 17 at 5:47
  • 1
    $\begingroup$ See this. Also, ELM should be removed from the page, as it's not new. Just look at the graphs representing ELM and FF/RBF. $\endgroup$ – Bullet51 Aug 17 at 10:16
  • $\begingroup$ I'm voting to close this question as off-topic because this is more suitable for datascience.stackexchange.com or ai.stackexchange.com $\endgroup$ – Andrea Ferretti Aug 20 at 16:04

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.