Search
Abstract
Class-incremental learning is becoming more popular as it helps models widen their applicability while not forgetting what they already know. A trend in this area is to use a mixture-of-expert technique, where different models work together to solve the task. However, the experts are usually trained all at once using whole task data, which makes them all prone to forgetting and increasing computational burden. To address this limitation, we introduce a novel approach named SEED. SEED selects only one, the most optimal expert for a considered task, and uses data from this task to fine-tune only this expert. For this purpose, each expert represents each class with a Gaussian distribution, and the optimal expert is selected based on the similarity of those distributions. Consequently, SEED increases diversity and heterogeneity within the experts while maintaining the high stability of this ensemble method. The extensive experiments demonstrate that SEED achieves state-of-the-art performance in exemplar-free settings across various scenarios, showing the potential of expert diversification through data in continual learning.
Authors (6)
-
Grzegorz Rypeść
-
Valeriya Khan
-
Tomasz Trzciński dr hab. inż.
-
Bartosz Zieliński dr hab inz
-
Bartłomiej Twardowski dr inż.
Cite as
Full text
Keywords
Details
- Category:
- Conference activity
- Type:
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language:
- English
- Publication year:
- 2024
- Bibliographic description:
- Rypeść G., Cygert S., Khan V., Trzciński T., Zieliński B., Twardowski B.: Divide and not forget: Ensemble of selectively trained experts in Continual Learning// / : , 2024,
- Sources of funding:
-
- finansowanie spoza PG
- Verified by:
- Gdańsk University of Technology
seen 48 times
Recommended for you
MagMax: Leveraging Model Merging for Seamless Continual Learning
- D. Marczak,
- B. Twardowski,
- T. Trzciński
- + 1 authors
Revisiting Supervision for Continual Representation Learning
- D. Marczak,
- S. Cygert,
- T. Trzciński
- + 1 authors
Category Adaptation Meets Projected Distillation in Generalized Continual Category Discovery
- G. Rypeść,
- D. Marczak,
- S. Cygert
- + 2 authors