Saved in:
Bibliographic Details
Main Authors: Cruttwell, Geoffrey S. H., Gavranovic, Bruno, Ghani, Neil, Wilson, Paul, Zanasi, Fabio
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.00408
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913292725780480
author Cruttwell, Geoffrey S. H.
Gavranovic, Bruno
Ghani, Neil
Wilson, Paul
Zanasi, Fabio
author_facet Cruttwell, Geoffrey S. H.
Gavranovic, Bruno
Ghani, Neil
Wilson, Paul
Zanasi, Fabio
contents We propose a categorical semantics for machine learning algorithms in terms of lenses, parametric maps, and reverse derivative categories. This foundation provides a powerful explanatory and unifying framework: it encompasses a variety of gradient descent algorithms such as ADAM, AdaGrad, and Nesterov momentum, as well as a variety of loss functions such as MSE and Softmax cross-entropy, and different architectures, shedding new light on their similarities and differences. Furthermore, our approach to learning has examples generalising beyond the familiar continuous domains (modelled in categories of smooth maps) and can be realised in the discrete setting of Boolean and polynomial circuits. We demonstrate the practical significance of our framework with an implementation in Python.
format Preprint
id arxiv_https___arxiv_org_abs_2404_00408
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Deep Learning with Parametric Lenses
Cruttwell, Geoffrey S. H.
Gavranovic, Bruno
Ghani, Neil
Wilson, Paul
Zanasi, Fabio
Machine Learning
Logic in Computer Science
We propose a categorical semantics for machine learning algorithms in terms of lenses, parametric maps, and reverse derivative categories. This foundation provides a powerful explanatory and unifying framework: it encompasses a variety of gradient descent algorithms such as ADAM, AdaGrad, and Nesterov momentum, as well as a variety of loss functions such as MSE and Softmax cross-entropy, and different architectures, shedding new light on their similarities and differences. Furthermore, our approach to learning has examples generalising beyond the familiar continuous domains (modelled in categories of smooth maps) and can be realised in the discrete setting of Boolean and polynomial circuits. We demonstrate the practical significance of our framework with an implementation in Python.
title Deep Learning with Parametric Lenses
topic Machine Learning
Logic in Computer Science
url https://arxiv.org/abs/2404.00408