An Investigation of Replay-based Approaches for Continual Learning
By Benedikt Bagus, Alexander Rainer Tassilo Gepperth. 2021
48 citations recorded by the source corpus.
Continual learning (CL) is a major challenge of machine learning (ML) and describes the ability to learn several tasks sequentially without catastrophic forgetting (CF). Recent works indicate that CL is a complex topic, even more so when real-world scenarios with multiple constraints are involved. Several solution classes have been proposed [1], of which socalled replay-based approaches seem very promising due to their simplicity and robustness. Such approaches store a subset of past samples in a dedicated memory for later processing: while this does not solve all problems, good results have been obtained. In this article, we empirically investigate replay-based approaches of continual learning and assess their potential for applications. Selected recent approaches as well as own proposals are compared on a common set of benchmarks, with a particular focus on assessing the performance of different sample selection strategies. We find that the impact of sample selection increases when a smaller number of samples is stored. Nevertheless, performance varies strongly between different replay approaches. Surprisingly, we find that the most naive rehearsal-based approaches that we propose here can outperform recent state-of-the-art methods.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Irina Rish
- Timothée Lesort
- Oleksiy Ostapenko
- Laurent Charlin
- Arthur Douillard
- P. Rodr'iguez
- Md Rifat Arefin
- Massimo Caccia
In Computer Vision and Pattern Recognition
- Timothée Lesort
- Irina Rish
- Laurent Charlin
- Oleksiy Ostapenko
- Arthur Douillard
- P. Rodr'iguez
- Md Rifat Arefin
- Massimo Caccia
Citations in the corpus
- Exploring Example Influence in Continual Learning (2022, neutral): “Rehearsal-based methods [29, 7, 22, 6, 1, 2, 28, 3, 25] tackle the challenge of SP dilemma by retaining a subset of old tasks in a stored memory buffer with bounded resources.”
- Continual Learning with Foundation Models: An Empirical Study of Latent Replay (2022, background): “It is important to mention that more elaborate replay strategies (Aljundi et al., 2019a;b; Bagus and Gepperth, 2021) can be used.”