Deep Learning as the Disciplined Construction of Tame Objects

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bareilles, Gilles, Gehret, Allen, Aspman, Johannes, Lepšová, Jana, Mareček, Jakub
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916072942206976
author Bareilles, Gilles
Gehret, Allen
Aspman, Johannes
Lepšová, Jana
Mareček, Jakub
author_facet Bareilles, Gilles
Gehret, Allen
Aspman, Johannes
Lepšová, Jana
Mareček, Jakub
contents One can see deep-learning models as compositions of functions within the so-called tame geometry. In this expository note, we give an overview of some topics at the interface of tame geometry (also known as o-minimality), optimization theory, and deep learning theory and practice. To do so, we gradually introduce the concepts and tools used to build convergence guarantees for stochastic gradient descent in a general nonsmooth nonconvex, but tame, setting. This illustrates some ways in which tame geometry is a natural mathematical framework for the study of AI systems, especially within Deep Learning.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18025
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep Learning as the Disciplined Construction of Tame Objects
Bareilles, Gilles
Gehret, Allen
Aspman, Johannes
Lepšová, Jana
Mareček, Jakub
Optimization and Control
Artificial Intelligence
Machine Learning
Logic
One can see deep-learning models as compositions of functions within the so-called tame geometry. In this expository note, we give an overview of some topics at the interface of tame geometry (also known as o-minimality), optimization theory, and deep learning theory and practice. To do so, we gradually introduce the concepts and tools used to build convergence guarantees for stochastic gradient descent in a general nonsmooth nonconvex, but tame, setting. This illustrates some ways in which tame geometry is a natural mathematical framework for the study of AI systems, especially within Deep Learning.
title Deep Learning as the Disciplined Construction of Tame Objects
topic Optimization and Control
Artificial Intelligence
Machine Learning
Logic
url https://arxiv.org/abs/2509.18025