Comments
There's unfortunately not much to read here yet...
Best of Machine Learning
Discover the best guides, books, papers and news in Machine Learning, once per week.
Follow the full discussion on Reddit.
I'm working on a open-ended project for my Masters using the Online Encyclopedia of Integer Sequences and attempting to upper bound future sequence values instead of predicting them. I've managed to get reasonable results using RNNs and Linear-Exponential (LINEX) loss but was wondering what other popular asymmetric loss functions for bounding there are or what general principles I should consider in constructing a loss function if my aim is to ensure all predicted bounds are deterministically (and not probabilistically) sufficient, if possible?
There's unfortunately not much to read here yet...
Ever having issues keeping up with everything that's going on in Machine Learning? That's where we help. We're sending out a weekly digest, highlighting the Best of Machine Learning.
Best of Machine Learning
Discover the best guides, books, papers and news in Machine Learning, once per week.
Join 800 others in curating the best parts of Machine Learning!
Just one more step. We use social logins to make sure that real humans are curating the list!
or contribute on GitHub directly.