Continuous Learning in Single-Incremental-Task Scenarios
read the original abstract
It was recently shown that architectural, regularization and rehearsal strategies can be used to train deep models sequentially on a number of disjoint tasks without forgetting previously acquired knowledge. However, these strategies are still unsatisfactory if the tasks are not disjoint but constitute a single incremental task (e.g., class-incremental learning). In this paper we point out the differences between multi-task and single-incremental-task scenarios and show that well-known approaches such as LWF, EWC and SI are not ideal for incremental task scenarios. A new approach, denoted as AR1, combining architectural and regularization strategies is then specifically proposed. AR1 overhead (in term of memory and computation) is very small thus making it suitable for online learning. When tested on CORe50 and iCIFAR-100, AR1 outperformed existing regularization strategies by a good margin.
This paper has not been read by Pith yet.
Forward citations
Cited by 1 Pith paper
-
Autoencoder-Based Incremental Class Learning without Retraining on Old Data
Autoencoder extracts class prototypes whose means enable metric classification in incremental learning, matching SOTA accuracy with lower memory overhead on CIFAR-100 and CUB-200-2011 via regularization to avoid forgetting.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.