Repository logo
 

IMPROVING FEATURE GENERALIZABILITY WITH MULTITASK LEARNING IN CLASS INCREMENTAL LEARNING

Accepted version
Peer-reviewed

Type

Conference Object

Change log

Authors

Ma, D 
Tang, CI 

Abstract

Many deep learning applications, like keyword spotting, require the incorporation of new concepts (classes) over time, referred to as Class Incremental Learning (CIL). The major challenge in CIL is catastrophic forgetting, i.e., preserving as much of the old knowledge as possible while learning new tasks. Various techniques, such as regularization, knowledge distillation, and the use of exemplars, have been proposed to resolve this issue. However, prior works primarily focus on the incremental learning step, while ignoring the optimization during the base model training. We hypothesize that a more transferable and generalizable feature representation from the base model would be beneficial to incremental learning. In this work, we adopt multitask learning during base model training to improve the feature generalizability. Specifically, instead of training a single model with all the base classes, we decompose the base classes into multiple subsets and regard each of them as a task. These tasks are trained concurrently and a shared feature extractor is obtained for incremental learning. We evaluate our approach on two datasets under various configurations. The results show that our approach enhances the average incremental learning accuracy by up to 5.5%, which enables more reliable and accurate keyword spotting over time. Moreover, the proposed approach can be combined with many existing techniques and provides additional performance gain.

Description

Keywords

Class Incremental Learning, Continual Learning, Multitask Learning, Keyword Spotting

Journal Title

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Conference Name

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Journal ISSN

1520-6149

Volume Title

2022-May

Publisher

IEEE
Sponsorship
European Commission Horizon 2020 (H2020) ERC (833296)