Neural Algorithmic Reasoning

TL;DR

This paper discusses neural algorithmic reasoning, a method for neural networks to execute algorithms, aiming to improve generalization and adapt algorithms to real-world problems beyond traditional capabilities.

Contribution

It introduces neural algorithmic reasoning as a new approach for neural networks to execute and adapt algorithms, enhancing their applicability and efficiency.

Findings

Neural networks can be trained to execute classical algorithms.

Neural algorithmic reasoning improves generalization over traditional deep learning.

Potential for neural networks to adapt algorithms to complex, real-world inputs.

Abstract

Algorithms have been fundamental to recent global technological advances and, in particular, they have been the cornerstone of technical advances in one field rapidly being applied to another. We argue that algorithms possess fundamentally different qualities to deep learning methods, and this strongly suggests that, were deep learning methods better able to mimic algorithms, generalisation of the sort seen with algorithms would become possible with deep learning -- something far out of the reach of current machine learning methods. Furthermore, by representing elements in a continuous space of learnt algorithms, neural networks are able to adapt known algorithms more closely to real-world problems, potentially finding more efficient and pragmatic solutions than those proposed by human computer scientists. Here we present neural algorithmic reasoning -- the art of building neural networks that are able to execute algorithmic computation -- and provide our opinion on its transformative potential for running classical algorithms on inputs previously considered inaccessible to them.