activity
20212023
most citedProbable Domain Generalization via Quantile Risk Minimization

17 citations · 22 across the 5 of their papers we have counts for

collaborators

8 papers

cs.CV2023

GIVT: Generative Infinite-Vocabulary Transformers

Michael Tschannen, Cian Eastwood, Fabian Mentzer

We introduce Generative Infinite-Vocabulary Transformers (GIVT) which generate vector sequences with real-valued entries, instead of discrete tokens from a finite vocabulary. To th…

cs.LG2023

Self-Supervised Disentanglement by Leveraging Structure in Data Augmentations

Cian Eastwood, Julius von Kügelgen, Linus Ericsson +4

Self-supervised representation learning often uses data augmentations to induce some invariance to "style" attributes of the data. However, with downstream tasks generally unknown…

cs.LG2023★ 2 cited

Spuriosity Didn't Kill the Classifier: Using Invariant Predictions to Harness Spurious Features

Cian Eastwood, Shashank Singh, Andrei Liviu Nicolicioiu +3

To avoid failures on out-of-distribution data, recent works have sought to extract features that have an invariant or stable relationship with the label across domains, discarding…

cs.LG2022★ 3 cited

DCI-ES: An Extended Disentanglement Framework with Connections to Identifiability

Cian Eastwood, Andrei Liviu Nicolicioiu, Julius von Kügelgen +4

In representation learning, a common approach is to seek representations which disentangle the underlying factors of variation. Eastwood & Williams (2018) proposed three metrics fo…

stat.ML2022★ 17 cited

Probable Domain Generalization via Quantile Risk Minimization

Cian Eastwood, Alexander Robey, Shashank Singh +4

Domain generalization (DG) seeks predictors which perform well on unseen test distributions by leveraging data drawn from multiple related training distributions or domains. To ach…

cs.CV2022

Align-Deform-Subtract: An Interventional Framework for Explaining Object Differences

Cian Eastwood, Li Nanbo, Christopher K. I. Williams

Given two object images, how can we explain their differences in terms of the underlying object properties? To address this question, we propose Align-Deform-Subtract (ADS) -- an i…