Neural Algorithmic Reasoning
This work addresses the challenge of improving generalization in deep learning for AI researchers and practitioners, though it appears incremental as it builds on existing algorithmic concepts.
The paper tackles the problem of deep learning's limited generalization by proposing neural algorithmic reasoning, which enables neural networks to mimic and adapt classical algorithms, potentially achieving generalization far beyond current machine learning methods.
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.