Neural Networks for Complex Data

TL;DR

This paper reviews recent advances in neural network models designed to handle complex data types like graphs and functions, highlighting progress made over the last decade.

Contribution

It provides a comprehensive summary of developments in neural networks tailored for complex data, emphasizing contributions from the SAMM team over the past decade.

Findings

Neural networks have evolved to process complex data structures.

Significant progress in neural network architectures for graphs and functions.

The review highlights key theoretical and practical advancements.

Abstract

Artificial neural networks are simple and efficient machine learning tools. Defined originally in the traditional setting of simple vector data, neural network models have evolved to address more and more difficulties of complex real world problems, ranging from time evolving data to sophisticated data structures such as graphs and functions. This paper summarizes advances on those themes from the last decade, with a focus on results obtained by members of the SAMM team of Universit\'e Paris 1