Search Results for author: Anna Sepliarskaia

Found 4 papers, 2 papers with code

On genuine invariance learning without weight-tying

1 code implementation7 Aug 2023 Artem Moskalev, Anna Sepliarskaia, Erik J. Bekkers, Arnold Smeulders

We demonstrate that even when a network learns to correctly classify samples on a group orbit, the underlying decision-making in such a model does not attain genuine invariance.

Decision Making

VC dimensions of group convolutional neural networks

no code implementations19 Dec 2022 Philipp Christian Petersen, Anna Sepliarskaia

We study the generalization capacity of group convolutional neural networks.

LieGG: Studying Learned Lie Group Generators

1 code implementation9 Oct 2022 Artem Moskalev, Anna Sepliarskaia, Ivan Sosnovik, Arnold Smeulders

Symmetries built into a neural network have appeared to be very beneficial for a wide range of tasks as it saves the data to learn them.

How to Not Measure Disentanglement

no code implementations12 Oct 2019 Anna Sepliarskaia, Julia Kiseleva, Maarten de Rijke

We conclude that existing metrics of disentanglement were created to reflect different characteristics of disentanglement and do not satisfy two basic desirable properties: (1) assign a high score to representations that are disentangled according to the definition; and (2) assign a low score to representations that are entangled according to the definition.

Disentanglement Transfer Learning +1

Cannot find the paper you are looking for? You can Submit a new open access paper.