Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

If I understand this, this model (dkm, "dynamic kernel matching") is to sequences as convolutional networks are to images?


Yeah, that's right. Just as a convolutional network picks a patch from an image using a max-pooling operation, dynamic kernel matching picks symbols in a sequence using a max-alignment operation




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: