Prime: A few primitives can boost robustness to common corruptions

A Modas, R Rade, G Ortiz-Jiménez… - … on Computer Vision, 2022 - Springer
Despite their impressive performance on image classification tasks, deep networks have a
hard time generalizing to unforeseen corruptions of their data. To fix this vulnerability, prior …

[HTML][HTML] Chevalley-Warning type results on abelian groups

E Aichinger, J Moosbauer - Journal of Algebra, 2021 - Elsevier
We develop a notion of degree for functions between two abelian groups that allows us to
generalize the Chevalley-Warning Theorems from fields to noncommutative rings or abelian …

Robustness and invariance properties of image classifiers

A Modas - arxiv preprint arxiv:2209.02408, 2022 - arxiv.org
Deep neural networks have achieved impressive results in many image classification tasks.
However, since their performance is usually measured in controlled settings, it is important …

Algorithms for Finite Near-rings and theirN-groups

F Binder, P Mayr - Journal of Symbolic Computation, 2001 - Elsevier
In this note, we present algorithms to deal with finite near-rings, the appropriate algebraic
structure to study non-linear functions on finite groups. Just as rings (of matrices) operate on …

Chevalley Warning Theorems on abelian groups/submitted by Jakob Moosbauer

J Moosbauer - 2019 - epub.jku.at
A theorem of Chevalley states that a system of polynomial equations over a finite field
cannot have exactly one solution if the number of variables is strictly greater than the sum of …

[PDF][PDF] A Bibliography of ISSAC (International Symposia on Symbolic and Algebraic Computation) Conference Papers

NHF Beebe - 2017 - ctan.math.utah.edu
A Bibliography of ISSAC (International Symposia on Symbolic and Algebraic Computation)
Conference Papers Page 1 A Bibliography of ISSAC (International Symposia on Symbolic …

[PDF][PDF] eines Doktors der Technischen Wissenschaften

DIJ Ecker - 2017 - epub.jku.at
Kurzfassung Formen der computervermittelten Kommunikation (CvK) sind allgegenwärtig
und beeinflussen unser Leben täglich. Facebook, Myspace, Skype, Twitter, WhatsApp und …