Abstract
The fields of Deep Learning and Abstract algebra are connected by the theory of symmetry groups. Convolutional Neural Network Layers are equiariant with respect to translations of an input image. Group Convolutional Networks are an generalization of regular CNN’s that are equivariant to the action of a general Group. I reproduce experiments that confirm an increase in performance of a CNN architecture by replacing convolutions with group convolutions.
Advisor
Kelvey, Robert
Department
Computer Science; Mathematics
Recommended Citation
Quinteros, Jorge, "Exploiting Symmetries in Training Convolutional Networks" (2025). Senior Independent Study Theses. Paper 11386.
https://openworks.wooster.edu/independentstudy/11386
Disciplines
Algebra | Artificial Intelligence and Robotics
Publication Date
2025
Degree Granted
Bachelor of Arts
Document Type
Senior Independent Study Thesis
© Copyright 2025 Jorge Quinteros