Practical and Consistent Estimation of f-Divergences

Autor:	Rubenstein, Paul K., Bousquet, Olivier, Djolonga, Josip, Riquelme, Carlos, Tolstikhin, Ilya
Rok vydání:	2019
Předmět:	Statistics - Machine Learning Computer Science - Information Theory Computer Science - Machine Learning
Zdroj:	33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada
Druh dokumentu:	Working Paper
Popis:	The estimation of an f-divergence between two probability distributions based on samples is a fundamental problem in statistics and machine learning. Most works study this problem under very weak assumptions, in which case it is provably hard. We consider the case of stronger structural assumptions that are commonly satisfied in modern machine learning, including representation learning and generative modelling with autoencoder architectures. Under these assumptions we propose and study an estimator that can be easily implemented, works well in high dimensions, and enjoys faster rates of convergence. We verify the behavior of our estimator empirically in both synthetic and real-data experiments, and discuss its direct implications for total correlation, entropy, and mutual information estimation. Comment: Accepted to the 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1905.11112 Zobrazit plný text záznamu View this record from Arxiv