2 Comments
User's avatar
spencer's avatar

potentially dumb question: is there a class techniques for loss functions that decrease the penalty of outliers? as opposed to robust techniques that just don't make the penalty as exteme. i realize the same could be done by removing outliers, and that this is only really a problem in gradient related optimization techniques, but could see it being helpful in high-dimension problems where outliers might not be as obvious so as to not update parameters so much cause of these outliers

Expand full comment
Avi Chawla's avatar

of course, there is a loss function. Huber loss helps. We covered it in this newsletter here: https://blog.dailydoseofds.com/p/11-powerful-techniques-to-supercharge

Expand full comment