Property of gradient of cross-entropy loss with kernel density estimation

Jump to navigation Jump to search

The thoughts on surfing & targeting is quite inspiring. And even if no data points are near within K size (hard case), that case is still valuable, since there may exist some data point just outside of the K size. And repeating the training process with new weight iteratively may eventually turn that case into an easy case ;) Are you doing something similar as well?

Xor (talk)08:27, 11 February 2022