Generative Expansion of Small Datasets: An Expansive Graph Approach

TL;DR

This paper presents an innovative graph-based generative model that creates large, diverse datasets from minimal samples, improving data augmentation for machine learning tasks.

Contribution

It introduces an expansive synthesis approach combining expander graphs, feature interpolation, and Koopman operators to generate high-quality, large-scale datasets from limited data.

Findings

Generated datasets enable classifiers to perform comparably to those trained on original data.

The model effectively preserves data distribution and feature relationships.

Results demonstrate potential for addressing data scarcity in machine learning.

Abstract

Limited data availability in machine learning significantly impacts performance and generalization. Traditional augmentation methods enhance moderately sufficient datasets. GANs struggle with convergence when generating diverse samples. Diffusion models, while effective, have high computational costs. We introduce an Expansive Synthesis model generating large-scale, information-rich datasets from minimal samples. It uses expander graph mappings and feature interpolation to preserve data distribution and feature relationships. The model leverages neural networks' non-linear latent space, captured by a Koopman operator, to create a linear feature space for dataset expansion. An autoencoder with self-attention layers and optimal transport refines distributional consistency. We validate by comparing classifiers trained on generated data to those trained on original datasets. Results show comparable performance, demonstrating the model's potential to augment training data effectively. This work advances data generation, addressing scarcity in machine learning applications.