Architect, Regularize and Replay (ARR): a Flexible Hybrid Approach for Continual Learning

TL;DR

ARR is a flexible hybrid method for continual learning that combines architectural design, regularization, and replay strategies, achieving state-of-the-art results across various data stream scenarios.

Contribution

It introduces a comprehensive hybrid approach that generalizes existing methods, enabling adaptable and efficient continual learning across diverse datasets.

Findings

Achieves state-of-the-art results in class-incremental learning

Generalizes well to real-world data streams

Balances efficiency and effectiveness through tunable parameters

Abstract

In recent years we have witnessed a renewed interest in machine learning methodologies, especially for deep representation learning, that could overcome basic i.i.d. assumptions and tackle non-stationary environments subject to various distributional shifts or sample selection biases. Within this context, several computational approaches based on architectural priors, regularizers and replay policies have been proposed with different degrees of success depending on the specific scenario in which they were developed and assessed. However, designing comprehensive hybrid solutions that can flexibly and generally be applied with tunable efficiency-effectiveness trade-offs still seems a distant goal. In this paper, we propose "Architect, Regularize and Replay" (ARR), an hybrid generalization of the renowned AR1 algorithm and its variants, that can achieve state-of-the-art results in classic scenarios (e.g. class-incremental learning) but also generalize to arbitrary data streams generated from real-world datasets such as CIFAR-100, CORe50 and ImageNet-1000.