Robust conditional probabilities

Yoav Wald, Amir Globerson

Research output: Contribution to journalConference articlepeer-review

Abstract

Conditional probabilities are a core concept in machine learning. For example, optimal prediction of a label Y given an input X corresponds to maximizing the conditional probability of Y given X. A common approach to inference tasks is learning a model of conditional probabilities. However, these models are often based on strong assumptions (e.g., log-linear models), and hence their estimate of conditional probabilities is not robust and is highly dependent on the validity of their assumptions. Here we propose a framework for reasoning about conditional probabilities without assuming anything about the underlying distributions, except knowledge of their second order marginals, which can be estimated from data. We show how this setting leads to guaranteed bounds on conditional probabilities, which can be calculated efficiently in a variety of settings, including structured-prediction. Finally, we apply them to semi-supervised deep learning, obtaining results competitive with variational autoencoders.

Original languageEnglish
Pages (from-to)6360-6369
Number of pages10
JournalAdvances in Neural Information Processing Systems
Volume2017-December
StatePublished - 2017
Event31st Annual Conference on Neural Information Processing Systems, NIPS 2017 - Long Beach, United States
Duration: 4 Dec 20179 Dec 2017

Fingerprint

Dive into the research topics of 'Robust conditional probabilities'. Together they form a unique fingerprint.

Cite this