BI-MAML: Balanced Incremental Approach for Meta Learning

TL;DR

BI-MAML introduces a balanced incremental meta-learning approach that effectively adapts to new tasks while preventing catastrophic forgetting, outperforming existing methods in accuracy and efficiency on benchmark datasets.

Contribution

The paper proposes BI-MAML, a novel meta-learning system with a balanced learning strategy that enables incremental adaptation to new tasks without forgetting old ones.

Findings

Outperforms state-of-the-art models in classification accuracy.

Achieves efficient adaptation with fewer training shots.

Demonstrates superior performance on benchmark image classification datasets.

Abstract

We present a novel Balanced Incremental Model Agnostic Meta Learning system (BI-MAML) for learning multiple tasks. Our method implements a meta-update rule to incrementally adapt its model to new tasks without forgetting old tasks. Such a capability is not possible in current state-of-the-art MAML approaches. These methods effectively adapt to new tasks, however, suffer from 'catastrophic forgetting' phenomena, in which new tasks that are streamed into the model degrade the performance of the model on previously learned tasks. Our system performs the meta-updates with only a few-shots and can successfully accomplish them. Our key idea for achieving this is the design of balanced learning strategy for the baseline model. The strategy sets the baseline model to perform equally well on various tasks and incorporates time efficiency. The balanced learning strategy enables BI-MAML to both outperform other state-of-the-art models in terms of classification accuracy for existing tasks and also accomplish efficient adaption to similar new tasks with less required shots. We evaluate BI-MAML by conducting comparisons on two common benchmark datasets with multiple number of image classification tasks. BI-MAML performance demonstrates advantages in both accuracy and efficiency.