Meta-Learning with Adaptive Hyperparameters

TL;DR

This paper introduces ALFA, a meta-network that adaptively generates hyperparameters for inner-loop optimization in MAML, significantly improving fast adaptation and outperforming traditional MAML even from random initialization.

Contribution

The paper proposes ALFA, a novel method that adaptively generates hyperparameters during fast adaptation, enhancing MAML's effectiveness especially when test tasks differ from training tasks.

Findings

ALFA outperforms MAML in few-shot learning tasks.

Fast adaptation with ALFA can surpass MAML even from random initialization.

Adaptive hyperparameters are crucial for effective meta-learning.

Abstract

Despite its popularity, several recent works question the effectiveness of MAML when test tasks are different from training tasks, thus suggesting various task-conditioned methodology to improve the initialization. Instead of searching for better task-aware initialization, we focus on a complementary factor in MAML framework, inner-loop optimization (or fast adaptation). Consequently, we propose a new weight update rule that greatly enhances the fast adaptation process. Specifically, we introduce a small meta-network that can adaptively generate per-step hyperparameters: learning rate and weight decay coefficients. The experimental results validate that the Adaptive Learning of hyperparameters for Fast Adaptation (ALFA) is the equally important ingredient that was often neglected in the recent few-shot learning approaches. Surprisingly, fast adaptation from random initialization with ALFA can already outperform MAML.