Remembering for the right reasons: Explanations reduce catastrophic forgetting

Remembering for the right reasons: Explanations reduce catastrophic forgetting

Abstract The goal of continual learning (CL) is to learn a sequence of tasks without suffering from the phenomenon of catastrophic forgetting. Previous work has shown that leveraging memory in the form of a replay buffer can reduce performance degradation on prior tasks. We hypothesize that forgetti...

Full description

Bibliographic Details
Main Authors: Sayna Ebrahimi, Suzanne Petryk, Akash Gokul, William Gan, Joseph E. Gonzalez, Marcus Rohrbach, Trevor Darrell
Format: Article
Language:English
Published: Wiley 2021-12-01
Series:Applied AI Letters
Subjects:
catastrophic forgetting
continual learning
explainable AI
lifelong learning
Online Access:https://doi.org/10.1002/ail2.44

Internet

https://doi.org/10.1002/ail2.44

Similar Items