Continual Learning with Pre-Trained Models: A Survey

TL;DR

This survey reviews recent advancements in continual learning that leverage pre-trained models, categorizing methodologies, comparing their strengths and weaknesses, and providing empirical evaluations to address fairness concerns.

Contribution

It offers a comprehensive categorization of PTM-based continual learning methods and includes an empirical study comparing state-of-the-art approaches.

Findings

Categorized PTM-based CL methods into three groups

Provided a comparative analysis of methodologies

Highlighted fairness issues in current evaluations

Abstract

Nowadays, real-world applications often face streaming data, which requires the learning system to absorb new knowledge as data evolves. Continual Learning (CL) aims to achieve this goal and meanwhile overcome the catastrophic forgetting of former knowledge when learning new ones. Typical CL methods build the model from scratch to grow with incoming data. However, the advent of the pre-trained model (PTM) era has sparked immense research interest, particularly in leveraging PTMs' robust representational capabilities. This paper presents a comprehensive survey of the latest advancements in PTM-based CL. We categorize existing methodologies into three distinct groups, providing a comparative analysis of their similarities, differences, and respective advantages and disadvantages. Additionally, we offer an empirical study contrasting various state-of-the-art methods to highlight concerns regarding fairness in comparisons. The source code to reproduce these evaluations is available at: https://github.com/sun-hailong/LAMDA-PILOT