3 - Seminar Meta Learning (SemMeL) - Benjamin Geissler - Meta-learning symmetries by reparameterization/ClipID:25222 previous clip next clip

Keywords: meta learning
Recording date 2020-11-30

Language

English

Organisational Unit

Friedrich-Alexander-Universität Erlangen-Nürnberg

Producer

Friedrich-Alexander-Universität Erlangen-Nürnberg

Today we have a presentation by Benjamin Geissler on "Meta-learning symmetries by reparameterization".

Abstract: Many successful deep learning architectures are equivariant to certain transformations in order to conserve parameters and improve generalization: most famously, convolution layers are equivariant to shifts of the input. This approach only works when practitioners know the symmetries of the task and can manually construct an architecture with the corresponding equivariances. Our goal is an approach for learning equivariances from data, without needing to design custom task-specific architectures. We present a method for learning and encoding equivariances into networks by learning corresponding parameter sharing patterns from data. Our method can provably represent equivariance-inducing parameter sharing for any finite group of symmetry transformations. Our experiments suggest that it can automatically learn to encode equivariances to common transformations used in image processing tasks. We provide our experiment code at this https URL.

https://arxiv.org/abs/2007.02933

More clips in this category "Friedrich-Alexander-Universität Erlangen-Nürnberg"

2024-02-07
IdM-login
protected  
2024-02-07
Studon
protected  
2024-02-12
IdM-login / Passwort / Studon
protected  
2024-02-07
IdM-login
protected  
2024-02-06
IdM-login / Studon
protected