Twisted Convolutional Networks (TCNs): Enhancing Feature Interactions for Non-Spatial Data Classification

TL;DR

Twisted Convolutional Networks (TCNs) introduce a novel architecture that explicitly models high-order feature interactions for improved classification of non-spatial data, outperforming traditional models across various domains.

Contribution

This paper presents TCNs, a new deep learning architecture that generalizes convolution to explicitly capture feature interactions using polynomial expansions.

Findings

TCNs outperform CNNs, ResNet, GNNs, DeepSets, and SVM on multiple datasets.

TCNs demonstrate improved training stability and generalization.

Statistically significant performance improvements are validated across diverse datasets.

Abstract

Twisted Convolutional Networks (TCNs) are proposed as a novel deep learning architecture for classifying one-dimensional data with arbitrary feature order and minimal spatial relationships. Unlike conventional Convolutional Neural Networks (CNNs) that rely on structured feature sequences, TCNs explicitly combine subsets of input features through theoretically grounded multiplicative and pairwise interaction mechanisms to create enriched representations. This feature combination strategy, formalized through polynomial feature expansions, captures high-order feature interactions that traditional convolutional approaches miss. We provide a comprehensive mathematical framework for TCNs, demonstrating how the twisted convolution operation generalizes standard convolutions while maintaining computational tractability. Through extensive experiments on five benchmark datasets from diverse domains (medical diagnostics, political science, synthetic data, chemometrics, and healthcare), we show that TCNs achieve statistically significant improvements over CNNs, Residual Networks (ResNet), Graph Neural Networks (GNNs), DeepSets, and Support Vector Machine (SVM). The performance gains are validated through statistical testing. TCNs also exhibit superior training stability and generalization capabilities, highlighting their robustness for non-spatial data classification tasks.