Task-Specific Preconditioner for Cross-Domain Few-Shot Learning

TL;DR

This paper introduces TSP, a novel task-specific preconditioning method for cross-domain few-shot learning that meta-learns domain-specific preconditioners to improve adaptation across diverse tasks.

Contribution

It proposes a new adaptation mechanism that meta-learns domain-specific preconditioners, enhancing gradient descent for better cross-domain few-shot learning performance.

Findings

Achieves state-of-the-art results on Meta-Dataset

Effectively adapts to diverse target tasks

Improves optimization efficiency in CDFSL

Abstract

Cross-Domain Few-Shot Learning~(CDFSL) methods typically parameterize models with task-agnostic and task-specific parameters. To adapt task-specific parameters, recent approaches have utilized fixed optimization strategies, despite their potential sub-optimality across varying domains or target tasks. To address this issue, we propose a novel adaptation mechanism called Task-Specific Preconditioned gradient descent~(TSP). Our method first meta-learns Domain-Specific Preconditioners~(DSPs) that capture the characteristics of each meta-training domain, which are then linearly combined using task-coefficients to form the Task-Specific Preconditioner. The preconditioner is applied to gradient descent, making the optimization adaptive to the target task. We constrain our preconditioners to be positive definite, guiding the preconditioned gradient toward the direction of steepest descent. Empirical evaluations on the Meta-Dataset show that TSP achieves state-of-the-art performance across diverse experimental scenarios.