UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Unsupervised representation learning with recognition-parametrised probabilistic models

Walker, william; Soulat, hugo; Yu, changmin; Sahani, Maneesh; (2023) Unsupervised representation learning with recognition-parametrised probabilistic models. In: Proceedings of the 26th International Conference on Artificial Intelligence and Statistics. Proceedings of Machine Learning Research: Valencia, Spain Proceedings of Machine Learning Research. (In press). Green open access

[thumbnail of walker-soulat-etal-2023-aistats.pdf]
Preview
Text
walker-soulat-etal-2023-aistats.pdf - Published Version

Download (6MB) | Preview

Abstract

We introduce a new approach to probabilistic unsupervised learning based on the recognitionparametrised model (RPM): a normalised semiparametric hypothesis class for joint distributions over observed and latent variables. Under the key assumption that observations are conditionally independent given latents, the RPM combines parametric prior and observation-conditioned latent distributions with non-parametric observation marginals. This approach leads to a flexible learnt recognition model capturing latent dependence between observations, without the need for an explicit, parametric generative model. The RPM admits exact maximum-likelihood learning for discrete latents, even for powerful neuralnetwork-based recognition. We develop effective approximations applicable in the continuouslatent case. Experiments demonstrate the effectiveness of the RPM on high-dimensional data, learning image classification from weak indirect supervision; direct image-level latent Dirichlet allocation; and recognition-parametrised Gaussian process factor analysis (RP-GPFA) applied to multi-factorial spatiotemporal datasets. The RPM provides a powerful framework to discover meaningful latent structure underlying observational data, a function critical to both animal and artificial intelligence.

Type: Proceedings paper
Title: Unsupervised representation learning with recognition-parametrised probabilistic models
Event: Artificial Intelligence and Statistics (AISTATS)
Open access status: An open access version is available from UCL Discovery
Publisher version: https://proceedings.mlr.press/
Language: English
Additional information: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences > Gatsby Computational Neurosci Unit
URI: https://discovery.ucl.ac.uk/id/eprint/10166301
Downloads since deposit
0Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item