InfoNCE is variational inference in a recognition parameterised model

no code implementations • 6 Jul 2021 • Laurence Aitchison, Stoil Ganev

In particular, we show that in the infinite sample limit, and for a particular choice of prior, the actual InfoNCE objective is equal to the ELBO (up to a constant); and the ELBO is equal to the marginal likelihood with a deterministic recognition model.

Bayesian Inference Self-Supervised Learning +1

Data augmentation in Bayesian neural networks and the cold posterior effect

no code implementations • 10 Jun 2021 • Seth Nabarro, Stoil Ganev, Adrià Garriga-Alonso, Vincent Fortuin, Mark van der Wilk, Laurence Aitchison

Here, we provide several approaches to developing principled Bayesian neural networks incorporating data augmentation.

Bayesian Inference Data Augmentation +1

Semi-supervised learning objectives as log-likelihoods in a generative model of data curation

no code implementations • 13 Aug 2020 • Stoil Ganev, Laurence Aitchison

We currently do not have an understanding of semi-supervised learning (SSL) objectives such as pseudo-labelling and entropy minimization as log-likelihoods, which precludes the development of e. g. Bayesian SSL.