Incrementally learning new classes with generative classification

Informations projet

GenClassifier4CL

N° de convention de subvention: 101067759

DOI

10.3030/101067759

Projet clôturé

Date de signature de la CE 23 Juin 2022

Date de début 1 Octobre 2022

Date de fin 30 Septembre 2024

Financé au titre de

• Marie Skłodowska-Curie Actions (MSCA)

Coût total

Aucune donnée

Contribution de l’UE

€ 191 760,00

Coordonné par

KATHOLIEKE UNIVERSITEIT LEUVEN
Belgium

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Documents, reports (2)

Communication, Dissemination & Outreach Plan

The plan describes the planned measures to maximize the impact of the project, including the dissemination and exploitation measures that are planned, and the target group(s) addressed. Regarding communication measures and public engagement strategy, the aim is to inform and reach out to society and show the activities performed, and the use and the benefits the project will have for citizens.

Career Development Plan

A Career Development Plan will be established jointly by the supervisor(s) and the researcher. In addition to research objectives, this plan will comprise the researcher's training and career needs, including training on transferable skills, teaching, planning for publications and participation in conferences and events aiming at opening science and research to citizens. The Plan can be updated when needed.

Data Management Plan (1)

Data Management Plan

The Data Management Plan describes the data management life cycle for all data sets that will be collected, processed or generated by the action. It is a document describing what data will be collected, processed or generated and following what methodology and standards, whether and how this data will be shared and/or made open, and how it will be curated and preserved.

Publications

Conference proceedings (4)

Continual Learning of Diffusion Models with Generative Distillation

Auteurs: Sergi Masip, Pau Rodriguez, Tinne Tuytelaars, Gido M van de Ven
Publié dans: Proceedings of The 3rd Conference on Lifelong Learning Agents (CoLLAs), 2024
Éditeur: To appear in PMLR

Prediction Error-based Classification for Class-Incremental Learning

Auteurs: Michał Zając, Tinne Tuytelaars, Gido M van de Ven
Publié dans: Proceedings of The Twelfth International Conference on Learning Representations - ICLR 2024, 2024
Éditeur: OpenReview.net

Two Complementary Perspectives to Continual Learning: Ask Not Only What to Optimize, But Also How

Auteurs: Timm Hess, Tinne Tuytelaars, Gido M van de Ven
Publié dans: Proceedings of the 1st ContinualAI Unconference, 2023, Numéro 249, 2024
Éditeur: PMLR

Continual evaluation for lifelong learning: Identifying the stability gap

Auteurs: Matthias De Lange, Gido M van de Ven, Tinne Tuytelaars
Publié dans: Proceedings of The Eleventh International Conference on Learning Representations - ICLR 2023, 2023
Éditeur: OpenReview.net

Peer reviewed articles (1)

Knowledge Accumulation in Continually Learned Representations and the Issue of Feature Forgetting

Auteurs: Timm Hess, Eli Verwimp, Gido M van de Ven, Tinne Tuytelaars
Publié dans: Transactions on Machine Learning Research (TMLR), 2024, ISSN 2835-8856
Éditeur: TMLR
DOI: 10.48550/arXiv.2304.00933

Other (1)

Continual Learning and Catastrophic Forgetting

Auteurs: Gido M. van de Ven, Nicholas Soures, Dhireesha Kudithipudi
Publié dans: arXiv, 2024
Éditeur: arXiv
DOI: 10.48550/ARXIV.2403.05175