Skip to main content
14 votes

Mathematics for machine learning

I think the answer depends on which structure you are looking to approximate, and, in what sense you want to approximate it. Below, you'll find a few contemporary references to help out :) Shallow ...
ABIM's user avatar
  • 5,405
14 votes

Structures of the space of neural networks

I would like to argue that the space of neural networks is a category with finite products, or more concretely a Lawvere theory. This expresses an important piece of structure, namely how neural ...
Tobias Fritz's user avatar
  • 6,406
13 votes

Is there any paper which summarizes the mathematical foundation of deep learning?

Mathematics of Deep Learning (2017) This tutorial will review recent work that aims to provide a mathematical justification for several properties of deep networks, such as global optimality, ...
Carlo Beenakker's user avatar
10 votes
Accepted

Structures of the space of neural networks

In information geometry, people study structures of Riemannian manifolds with dual affine connections on sets of neural networks. (The metric measures how close neural networks are in their input-...
Matthias Wendt's user avatar
9 votes

Neural networks over gadgets other than $\mathbb{R}$

Studies of neural networks that are more general than $\mathbb{R}^n\mapsto\mathbb{R}^k$ include Deep Complex Networks (2017) Deep Quaternion Networks (2017) P-Adic Neural Networks (2004) Deep ...
Carlo Beenakker's user avatar
8 votes
Accepted

Difference between deep neural networks and expectation maximization algorithm

A high level view for neural networks is that they are just heuristics for approximating multi-variate functions. You pick a class of parametrizable functions and you optimize their parameters. ...
Jim's user avatar
  • 330
8 votes

Theoretical results on neural networks

Your question is a bit too broad, but here is something you may want to read, if you are interested in the Mathematical Analysis of Deep Learning: The Modern Mathematics of Deep Learning, by Berner et ...
7 votes

Abstract mathematical concepts/tools appeared in machine learning research

Probably, one the most striking is the "UMAP" (Uniform manifold approximation and projection) - a method of dimensional reduction in machine learning. The authors of the method use CATEGORY THEORY ...
7 votes

Mathematics for machine learning

I believe Ian Goodfellow and Yoshua Bengio's Deep Learning book covers the basics and also how you would use it for research. The chapters are also available online for free.
Dendi Suhubdy's user avatar
7 votes

Deep learning / Deep neural nets for mathematician

Since this question got bumped up to the front page somehow, I'm taking the liberty to suggest a partial introduction to the "Math of Deep Learning" given in the following article: The ...
7 votes

Using a poset or directed graph as input for a neural network

This is most definitely NOT the right spot to ask such a question, thought it is a good one. I happen to dabble precisely with these things in these days, for my own work and research, so I think I ...
Mirco A. Mannucci's user avatar
7 votes

Theoretical results on neural networks

ICLR 2021 has contributions that could qualify as "rigorous results", one you may like is Minimum Width for Universal Approximation. The universal approximation property of width-bounded ...
6 votes

Theoretical results on neural networks

The Representer Theorem by Michael Unser has recently unveiled explicit connections between deep NNs (using ReLUs as nonlinearities I believe) and splines. One core idea is that both the linear ...
5 votes

Deep learning / Deep neural nets for mathematician

Note: The following is an answer to this post but everything I posted there applies equally and fully here. Let me also comment shortly, that I also found the theory of DNNs difficult to enter since a ...
5 votes

Deep learning / Deep neural nets for mathematician

I just found this paper-https://arxiv.org/pdf/1801.05894.pdf, which introduces deep learning in a mathematically sound manner, specially for computations of backpropagation etc. As a mathematician who ...
4 votes

Universal approximation theorem for whole $\mathbb{R}^d$

I do not think a universal approximation theorem on all of $\mathbb{R}^d$ is possible with the uniform norm. In $L^p$ for $p < \infty$ there may be hope in some cases. Let us first look at the ...
pcp's user avatar
  • 141
4 votes

Mathematics for machine learning

As a deep learning practitioner with mathematical background I was yearning to have some satisfying mathematical framework of what I do in my every day job. In my opinion, very well fitted ...
Fallen Apart's user avatar
  • 1,615
4 votes

Deep learning / Deep neural nets for mathematician

There's an ongoing course taught by Elchanan Mossel at MIT that you might find helpful. It really focuses on the things we can actually prove about deep learning, which may be mathematically appealing ...
3 votes

Deep learning / Deep neural nets for mathematician

Eldad Haber at the University of British Columbia is exploring connections between neural networks, and dynamical systems: (2018) Deep Neural Networks Motivated by Partial Differential Equations (...
2 votes

Uniform Lipschitz function approximation by shallow neural networks

Maybe you can check Theorem 4 in Poggio et al. "Why and When Can Deep-but Not Shallow-Networks Avoid the Curse of Dimensionality: A Review"
Luca DGA's user avatar
2 votes

Deep learning / Deep neural nets for mathematician

I'd say that deep learning (from a mathematician's perspective) is a HOT MESS. People are jumping through hoops trying to increase accuracy (sometimes just by decimals) introducing all sorts of stuff ...
2 votes

Deep learning / Deep neural nets for mathematician

The original question was asked in 2015. So I believe it is appropriate to include surveys with mathematical flavor on more recent/advanced topics in neural networks and deep learning. A Mathematical ...
1 vote

How sensitive are Neural Networks to weight change?

Output of the ReLU network is $$v = \sum_{ij} X_i A_{ij} w^{(1)}_{ij} \cdots w^{(L)}_{ij} $$ where $i$ is the input $j$ is the path and $A_{ij}$ is $1$ if the path is open and $0$ otherwise. Now if ...
Mikko Pitkonen's user avatar
1 vote

Mathematics for machine learning

I recently wrote an answer to a related question, about math research that can enhance machine learning. But, part of what I wrote is also related to a learning resource that might help someone ...
David White's user avatar
  • 30.3k

Only top scored, non community-wiki answers of a minimum length are eligible