Saving for the future: Enhancing generalization via partial logic regularization

TL;DR

This paper introduces PL-Reg, a partial logic regularization method that enhances model generalization to unknown classes in visual classification tasks by reserving space for undefined logical formulas.

Contribution

It proposes a novel partial-logic regularization technique that improves adaptability to unknown classes and demonstrates its effectiveness through extensive experiments.

Findings

PL-Reg improves generalization to unknown classes.

The method outperforms existing approaches in various tasks.

Partial logic provides better flexibility for class discovery.

Abstract

Generalization remains a significant challenge in visual classification tasks, particularly in handling unknown classes in real-world applications. Existing research focuses on the class discovery paradigm, which tends to favor known classes, and the incremental learning paradigm, which suffers from catastrophic forgetting. Recent approaches such as the L-Reg technique employ logic-based regularization to enhance generalization but are bound by the necessity of fully defined logical formulas, limiting flexibility for unknown classes. This paper introduces PL-Reg, a novel partial-logic regularization term that allows models to reserve space for undefined logic formulas, improving adaptability to unknown classes. Specifically, we formally demonstrate that tasks involving unknown classes can be effectively explained using partial logic. We also prove that methods based on partial logic lead to improved generalization. We validate PL-Reg through extensive experiments on Generalized Category Discovery, Multi-Domain Generalized Category Discovery, and long-tailed Class Incremental Learning tasks, demonstrating consistent performance improvements. Our results highlight the effectiveness of partial logic in tackling challenges related to unknown classes.