SPRINT: Semi-supervised Prototypical Representation for Few-Shot Class-Incremental Tabular Learning

TL;DR

SPRINT is a novel semi-supervised framework designed for few-shot class-incremental learning in tabular data, effectively utilizing unlabeled data and low storage to improve adaptability and accuracy across diverse real-world domains.

Contribution

It introduces the first FSCIL method for tabular data, employing confidence-based pseudo-labeling and a mixed training strategy to enhance class representations and retain knowledge.

Findings

Achieves 77.37% accuracy in 5-shot settings.

Outperforms previous methods by 4.45% on average.

Demonstrates robustness across cybersecurity, healthcare, and ecological benchmarks.

Abstract

Real-world systems must continuously adapt to novel concepts from limited data without forgetting previously acquired knowledge. While Few-Shot Class-Incremental Learning (FSCIL) is established in computer vision, its application to tabular domains remains largely unexplored. Unlike images, tabular streams (e.g., logs, sensors) offer abundant unlabeled data, a scarcity of expert annotations and negligible storage costs, features ignored by existing vision-based methods that rely on restrictive buffers. We introduce SPRINT, the first FSCIL framework tailored for tabular distributions. SPRINT introduces a mixed episodic training strategy that leverages confidence-based pseudo-labeling to enrich novel class representations and exploits low storage costs to retain base class history. Extensive evaluation across six diverse benchmarks spanning cybersecurity, healthcare, and ecological domains, demonstrates SPRINT's cross-domain robustness. It achieves a state-of-the-art average accuracy of 77.37% (5-shot), outperforming the strongest incremental baseline by 4.45%.