Semi-Supervised Learning with Normalizing Flows

12/30/2019
by   Pavel Izmailov, et al.
13

Normalizing flows transform a latent distribution through an invertible neural network for a flexible and pleasingly simple approach to generative modelling, while preserving an exact likelihood. We propose FlowGMM, an end-to-end approach to generative semi supervised learning with normalizing flows, using a latent Gaussian mixture model. FlowGMM is distinct in its simplicity, unified treatment of labelled and unlabelled data with an exact likelihood, interpretability, and broad applicability beyond image data. We show promising results on a wide range of applications, including AG-News and Yahoo Answers text data, tabular data, and semi-supervised image classification. We also show that FlowGMM can discover interpretable structure, provide real-time optimization-free feature visualizations, and specify well calibrated predictive distributions.

READ FULL TEXT

page 4

page 8

page 13

page 14

page 15

07/06/2020

Learning the Prediction Distribution for Semi-Supervised Learning with Normalising Flows

As data volumes continue to grow, the labelling process increasingly bec...
08/13/2020

A statistical theory of semi-supervised learning

We currently lack a solid statistical understanding of semi-supervised l...
05/01/2019

Semi-Conditional Normalizing Flows for Semi-Supervised Learning

This paper proposes a semi-conditional normalizing flow model for semi-s...
06/16/2021

A Survey on Semi-Supervised Learning for Delayed Partially Labelled Data Streams

Unlabelled data appear in many domains and are particularly relevant to ...
11/01/2019

Variational Autoencoders for Generative Modelling of Water Cherenkov Detectors

Matter-antimatter asymmetry is one of the major unsolved problems in phy...
04/04/2018

Generative Visual Rationales

Interpretability and small labelled datasets are key issues in the pract...
02/03/2015

Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Probe and Learn Neural Networks

In this paper, we propose a novel model for high-dimensional data, calle...

Code Repositories