# Toward a Characterization of Loss Functions for Distribution Learning

Nika HaghtalabCameron MuscoBo Waggoner

In this work we study loss functions for learning and evaluating probability distributions over large discrete domains. Unlike classification or regression where a wide variety of loss functions are used, in the distribution learning and density estimation literature, very few losses outside the dominant $log\ loss$ are applied... (read more)

PDF Abstract NeurIPS 2019 PDF NeurIPS 2019 Abstract

No code implementations yet. Submit your code now