Equivariant bifurcation, quadratic equivariants, and symmetry breaking for the standard representation of S k

From MaRDI portal
Publication:5865779

DOI10.1088/1361-6544/AC619FzbMATH Open1505.37041arXiv2107.02422OpenAlexW4287084359MaRDI QIDQ5865779FDOQ5865779


Authors: Yossi Arjevani, Michael Field Edit this on Wikidata


Publication date: 10 June 2022

Published in: Nonlinearity (Search for Journal in Brave)

Abstract: Motivated by questions originating from the study of a class of shallow student-teacher neural networks, methods are developed for the analysis of spurious minima in classes of gradient equivariant dynamics related to neural nets. In the symmetric case, methods depend on the generic equivariant bifurcation theory of irreducible representations of the symmetric group on n symbols, Sn; in particular, the standard representation of Sn. It is shown that spurious minima do not arise from spontaneous symmetry breaking but rather through a complex deformation of the landscape geometry that can be encoded by a generic Sn-equivariant bifurcation. We describe minimal models for forced symmetry breaking that give a lower bound on the dynamic complexity involved in the creation of spurious minima when there is no symmetry. Results on generic bifurcation when there are quadratic equivariants are also proved; this work extends and clarifies results of Ihrig & Golubitsky and Chossat, Lauterback & Melbourne on the instability of solutions when there are quadratic equivariants.


Full work available at URL: https://arxiv.org/abs/2107.02422




Recommendations




Cites Work


Cited In (1)





This page was built for publication: Equivariant bifurcation, quadratic equivariants, and symmetry breaking for the standard representation of S k

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5865779)