Gauge equivariant convolutional neural networks

Examensarbete för masterexamen

Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.12380/301431
Download file(s):
File Description SizeFormat 
Master_Thesis_Oscar_Carlsson.pdf1.26 MBAdobe PDFView/Open
Bibliographical item details
FieldValue
Type: Examensarbete för masterexamen
Title: Gauge equivariant convolutional neural networks
Authors: Carlsson, Oscar
Abstract: In this thesis we present a review of the current theory of group and gauge equivariant convolutional neural networks on homogeneous spaces and general smooth manifolds, with focus on the latter, formulated from a mathematical viewpoint. We also provide a new interpretation of layers in neural networks as maps between associated bundles. Furthermore we discuss the implementation of simple convolutional neural networks invariant under 90 rotations and reflections, build such networks, and test them to show the effect of the invariant construction. This testing shows that the addition of the group invariant structure allows the network to efficiently classify transformed data while only training on untransformed data.
Keywords: Convolutional neural networks;machine learning;manifolds;group;gauge;Python;Tensorflow;Keras
Issue Date: 2020
Publisher: Chalmers tekniska högskola / Institutionen för fysik
URI: https://hdl.handle.net/20.500.12380/301431
Collection:Examensarbeten för masterexamen // Master Theses



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.