hh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Multimodal meta-learning through meta-learned task representations
Halmstad University, School of Information Technology, Center for Applied Intelligent Systems Research (CAISR).ORCID iD: 0000-0003-0185-5038
Halmstad University, School of Information Technology, Center for Applied Intelligent Systems Research (CAISR).ORCID iD: 0000-0002-2859-6155
Halmstad University, School of Information Technology, Center for Applied Intelligent Systems Research (CAISR).ORCID iD: 0000-0001-5163-2997
2024 (English)In: Neural Computing & Applications, ISSN 0941-0643, E-ISSN 1433-3058, Vol. 36, no 15, p. 8519-8529Article in journal (Refereed) Published
Abstract [en]

Few-shot meta-learning involves training a model on multiple tasks to enable it to efficiently adapt to new, previously unseen tasks with only a limited number of samples. However, current meta-learning methods assume that all tasks are closely related and belong to a common domain, whereas in practice, tasks can be highly diverse and originate from multiple domains, resulting in a multimodal task distribution. This poses a challenge for existing methods as they struggle to learn a shared representation that can be easily adapted to all tasks within the distribution. To address this challenge, we propose a meta-learning framework that can handle multimodal task distributions by conditioning the model on the current task, resulting in a faster adaptation. Our proposed method learns to encode each task and generate task embeddings that modulate the model’s activations. The resulting modulated model becomes specialized for the current task and leads to more effective adaptation. Our framework is designed to work in a realistic setting where the mode from which a task is sampled is unknown. Nonetheless, we also explore the possibility of incorporating auxiliary information, such as the task-mode-label, to further enhance the performance of our method if such information is available. We evaluate our proposed framework on various few-shot regression and image classification tasks, demonstrating its superiority over other state-of-the-art meta-learning methods. The results highlight the benefits of learning to embed task-specific information in the model to guide the adaptation when tasks are sampled from a multimodal distribution. © The Author(s) 2024.

Place, publisher, year, edition, pages
London: Springer, 2024. Vol. 36, no 15, p. 8519-8529
Keywords [en]
Meta-learning, Few-shot learning, Transfer learning, Task representation, Multimodal distribution
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:hh:diva-52924DOI: 10.1007/s00521-024-09540-1ISI: 001167510900006Scopus ID: 2-s2.0-85187241643OAI: oai:DiVA.org:hh-52924DiVA, id: diva2:1846103
Funder
Halmstad UniversityKnowledge FoundationAvailable from: 2024-03-21 Created: 2024-03-21 Last updated: 2025-01-07Bibliographically approved
In thesis
1. Advancing Meta-Learning for Enhanced Generalization Across Diverse Tasks
Open this publication in new window or tab >>Advancing Meta-Learning for Enhanced Generalization Across Diverse Tasks
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Meta-learning, or learning to learn, is a rapidly evolving area in machine learning that aims to enhance the adaptability and efficiency of learning algorithms. Inspired by the human ability to learn new concepts from limited examples and quickly adapt to unforeseen situations, meta-learning leverages prior experience to prepare models for fast adaptation to new tasks. Unlike traditional machine learning systems, where models are trained for specific tasks, meta-learning frameworks enable models to acquire generalized knowledge during training and efficiently learn new tasks during inference. This ability to generalize from past experiences to new tasks makes meta-learning a key focus in advancing artificial intelligence, offering the potential to create more flexible and efficient AI systems capable of performing well with minimal data.

In this thesis, we begin by formally defining the meta-learning framework, establishing clear terminology, and synthesizing existing work in a comprehensive survey paper. Building on this foundation, we demonstrate how meta-learning can be integrated into various fields to enhance model performance and extend capabilities to few-shot learning scenarios. We show how meta-learning can significantly improve the accuracy and efficiency of transferring knowledge across domains in domain adaptation. In scenarios involving a multimodal distribution of tasks, we develop methods that efficiently learn from and adapt to a wide variety of tasks drawn from different modes within the distribution, ensuring effective adaptation across diverse domains. Our work on personalized federated learning highlights meta-learning's potential to tailor federated learning processes to individual user needs while maintaining privacy and data security. Additionally, we address the challenges of continual learning by developing models that continuously integrate new information without forgetting previously acquired knowledge. For time series data analysis, we present meta-learning strategies that automatically learn optimal augmentation techniques, enhancing model predictions and offering robust solutions for real-world applications. Lastly, our pioneering research on unsupervised meta-learning via in-context learning explores innovative approaches for constructing tasks and learning effectively from unlabeled data.

Overall, the contributions of this thesis emphasize the potential of meta-learning techniques to improve performance across diverse research areas and demonstrate how advancements in one area can benefit the field as a whole.

Place, publisher, year, edition, pages
Halmstad: Halmstad University Press, 2025. p. 46
Series
Halmstad University Dissertations ; 127
Keywords
Meta-learning, Few-shot learning, Domain adaptation, Federated learning, Continual learning, Unsupervised learning, In-context learning
National Category
Computer Sciences
Identifiers
urn:nbn:se:hh:diva-55147 (URN)978-91-89587-71-7 (ISBN)978-91-89587-70-0 (ISBN)
Public defence
2025-02-03, S1022, Kristian IV:s väg 3, 30118, Halmstad, Halmstad, 13:00 (English)
Opponent
Supervisors
Available from: 2025-01-08 Created: 2025-01-07 Last updated: 2025-01-08Bibliographically approved

Open Access in DiVA

Fulltext(903 kB)54 downloads
File information
File name FULLTEXT02.pdfFile size 903 kBChecksum SHA-512
2741cf0417c1c9b599b814e97ed4e4bf5878deceb4f448f40ab5ce69d09bb27e532324d3522ee3e3146bac051fa94360d67f47c2d4e1355e5094b34938561c09
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopus

Authority records

Vettoruzzo, AnnaBouguelia, Mohamed-RafikRögnvaldsson, Thorsteinn

Search in DiVA

By author/editor
Vettoruzzo, AnnaBouguelia, Mohamed-RafikRögnvaldsson, Thorsteinn
By organisation
Center for Applied Intelligent Systems Research (CAISR)
In the same journal
Neural Computing & Applications
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 387 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 633 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf