LGAISep 21, 2022

Periodic Extrapolative Generalisation in Neural Networks

arXiv:2209.10280v14 citationsh-index: 25
Originality Incremental advance
AI Analysis

This addresses a fundamental challenge in neural network generalization for researchers in machine learning, though it is incremental in evaluating existing methods.

The paper tackles the problem of neural networks learning periodicity for extrapolative generalization, finding that traditional sequential models outperform novel periodic architectures, with population-based training achieving the best results.

The learning of the simplest possible computational pattern -- periodicity -- is an open problem in the research of strong generalisation in neural networks. We formalise the problem of extrapolative generalisation for periodic signals and systematically investigate the generalisation abilities of classical, population-based, and recently proposed periodic architectures on a set of benchmarking tasks. We find that periodic and "snake" activation functions consistently fail at periodic extrapolation, regardless of the trainability of their periodicity parameters. Further, our results show that traditional sequential models still outperform the novel architectures designed specifically for extrapolation, and that these are in turn trumped by population-based training. We make our benchmarking and evaluation toolkit, PerKit, available and easily accessible to facilitate future work in the area.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes