Latent Space Representations of Neural Algorithmic Reasoners
Abstract
Neural Algorithmic Reasoning (NAR) is a research area focused on designing neural architectures that can reliably capture classical computation, usually by learning to execute algorithms. A typical approach is to rely on Graph Neural Network (GNN) architectures, which encode inputs in high-dimensional latent spaces that are repeatedly transformed during the execution of the algorithm. In this work we perform a detailed analysis of the structure of the latent space induced by the GNN when executing algorithms. We identify two possible failure modes: (i) loss of resolution, making it hard to distinguish similar values; (ii) inability to deal with values outside the range observed during training. We propose to solve the first issue by relying on a softmax aggregator, and propose to decay the latent space in order to deal with out-of-range values. We show that these changes lead to improvements on the majority of algorithms in the standard CLRS-30 benchmark when using the state-of-the-art Triplet-GMPNN processor. Our code is available at https://github.com/mirjanic/nar-latent-spaces
Cite
@article{arxiv.2307.08874,
title = {Latent Space Representations of Neural Algorithmic Reasoners},
author = {Vladimir V. Mirjanić and Razvan Pascanu and Petar Veličković},
journal= {arXiv preprint arXiv:2307.08874},
year = {2024}
}
Comments
24 pages, 19 figures; Accepted at the Second Learning on Graphs Conference (LoG 2023); updated layout, reorganized content, added journal reference