3 ms·
What's the difference with MSE loss?
by qinjian623 5y ago
What's the difference with MSE loss?
- nerdponx 5y agoBrier score is MSE applied to predicted probabilities. One is a loss function for "regression" type problems, the other is a proper scoring rule.
- master_yoda_1 5y agoNo difference this is exactly same as brier score. MSE is the KL divergence between ground truth and true prediction, assuming a gaussian error distribution. We use MSE as loss because we try to minimize KL divergence (again assuming gaussian error distribution). The article is very shallow, I am surprise it comes on HN front page.
- nerdponx 5y agoGreat point about KL divergence and assumptions about error distribution. This kind of thing is what I think is missing from a lot of data science education.
- datarecipes 5y agoAgreed. It would be great to hear your views on some of the key gaps in modern data science curricula that could be covered in the blog - would you be able to drop me a line at datarecipes@pm.me? Thanks!
- datarecipes 5y agoThanks for your comment. I agree that the post lacks depth, but it was intended to be a gentle article accessible to a general audience, so they can start applying it in practice in their day to day lives. I would, however, really love to hear your views on what might be a more rigorous treatment of similar topics that can be introduced in an accessible way - would you be able to drop me a line at datarecipes@pm.me? Thanks!