Papers › Understanding Catastrophic Forgetting and Remembering in Continual Learning with...

Understanding Catastrophic Forgetting and Remembering in Continual Learning with Optimal Relevance Mapping

22 Feb 2021arXiv:2102.11343archive 2025-07-28

Prakhar Kaushik, Alex Gain, Adam Kortylewski, Alan Yuille

Catastrophic forgetting in neural networks is a significant problem for continual learning. A majority of the current methods replay previous data during training, which violates the constraints of an ideal continual learning system. Additionally, current approaches that deal with forgetting ignore the problem of catastrophic remembering, i.e. the worsening ability to discriminate between data from different tasks. In our work, we introduce Relevance Mapping Networks (RMNs) which are inspired by the Optimal Overlap Hypothesis. The mappings reflects the relevance of the weights for the task at hand by assigning large weights to essential parameters. We show that RMNs learn an optimized representational overlap that overcomes the twin problem of catastrophic forgetting and remembering. Our approach achieves state-of-the-art performance across all common continual learning datasets, even significantly outperforming data replay methods while not violating the constraints for an ideal continual learning system. Moreover, RMNs retain the ability to detect data from new tasks in an unsupervised manner, thus proving their resilience against catastrophic remembering.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Continual Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Continual Learning Cifar100 (10 tasks) RMN (Resnet) Average Accuracy 84.9 #3 of 5 Archive leaderboard report
Continual Learning Cifar100 (20 tasks) RMN Average Accuracy 81 #4 of 9 Archive leaderboard report
Continual Learning ImageNet-50 (5 tasks) RMN Accuracy 68.1 #1 of 5 Archive leaderboard report
Continual Learning Permuted MNIST RMN Average Accuracy 97.988 #1 of 3 Archive leaderboard report
Continual Learning Permuted MNIST RMN MLP Hidden Layers-width 2-100 #1 of 3 Archive leaderboard report
Continual Learning Permuted MNIST RMN Pretrained/Transfer Learning No #1 of 3 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections