Polyra Swarms: A Shape-Based Approach to Machine Learning

TL;DR

Polyra Swarms is a new machine-learning approach that models shapes instead of functions, offering advantages like low bias, improved generalization, and transparency, especially for anomaly detection tasks.

Contribution

The paper introduces Polyra Swarms, a shape-based learning method with an automated abstraction mechanism, providing a novel alternative to neural networks with distinct strengths.

Findings

Polyra Swarms can outperform neural networks in anomaly detection.

The abstraction mechanism simplifies models while maintaining performance.

Polyra Swarms offer a new research direction with different strengths and limitations.

Abstract

We propose Polyra Swarms, a novel machine-learning approach that approximates shapes instead of functions. Our method enables general-purpose learning with very low bias. In particular, we show that depending on the task, Polyra Swarms can be preferable compared to neural networks, especially for tasks like anomaly detection. We further introduce an automated abstraction mechanism that simplifies the complexity of a Polyra Swarm significantly, enhancing both their generalization and transparency. Since Polyra Swarms operate on fundamentally different principles than neural networks, they open up new research directions with distinct strengths and limitations.