A Benchmark and Empirical Analysis for Replay Strategies in Continual Learning
This work addresses the problem of catastrophic forgetting in deep neural networks for continual learning applications, offering an incremental analysis of existing replay methods.
The paper conducted a comprehensive evaluation of memory replay methods for mitigating catastrophic forgetting in continual learning, testing efficiency, performance, and scalability across multiple datasets and domains, and provided a practical solution for selecting replay strategies based on data distributions.
With the capacity of continual learning, humans can continuously acquire knowledge throughout their lifespan. However, computational systems are not, in general, capable of learning tasks sequentially. This long-standing challenge for deep neural networks (DNNs) is called catastrophic forgetting. Multiple solutions have been proposed to overcome this limitation. This paper makes an in-depth evaluation of the memory replay methods, exploring the efficiency, performance, and scalability of various sampling strategies when selecting replay data. All experiments are conducted on multiple datasets under various domains. Finally, a practical solution for selecting replay methods for various data distributions is provided.