6

[2209.11142] A Generalist Neural Algorithmic Learner

 1 year ago
source link: https://arxiv.org/abs/2209.11142
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
neoserver,ios ssh client

[Submitted on 22 Sep 2022 (v1), last revised 3 Dec 2022 (this version, v2)]

A Generalist Neural Algorithmic Learner

Download PDF

The cornerstone of neural algorithmic reasoning is the ability to solve algorithmic tasks, especially in a way that generalises out of distribution. While recent years have seen a surge in methodological improvements in this area, they mostly focused on building specialist models. Specialist models are capable of learning to neurally execute either only one algorithm or a collection of algorithms with identical control-flow backbone. Here, instead, we focus on constructing a generalist neural algorithmic learner -- a single graph neural network processor capable of learning to execute a wide range of algorithms, such as sorting, searching, dynamic programming, path-finding and geometry. We leverage the CLRS benchmark to empirically show that, much like recent successes in the domain of perception, generalist algorithmic learners can be built by "incorporating" knowledge. That is, it is possible to effectively learn algorithms in a multi-task manner, so long as we can learn to execute them well in a single-task regime. Motivated by this, we present a series of improvements to the input representation, training regime and processor architecture over CLRS, improving average single-task performance by over 20% from prior art. We then conduct a thorough ablation of multi-task learners leveraging these improvements. Our results demonstrate a generalist learner that effectively incorporates knowledge captured by specialist models.

Comments: To appear at LoG 2022 (Spotlight talk). 23 pages, 11 figures
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
Cite as: arXiv:2209.11142 [cs.LG]
  (or arXiv:2209.11142v2 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2209.11142

About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK