Call for Papers

In this international workshop, we aim at bringing together researchers across the ML community to present and discuss recent results in reliable learning across domains, fostering new connections between theory and practical methods, and identifying solutions targeting different modalities (images, videos, language, and more) and application areas.

Paper submissions are now open. Visit: https://cmt3.research.microsoft.com/ECMLPKDDworkshop2023/Track/42/Submission/Create

The topics that are of special interest (though submissions are not limited to these) are:

• Transfer Learning

• Domain Generalization (DG)

  • Data Augmentation Approaches

  • Adversarial Approaches

  • Regularization Approaches

• Domain Adaptation (DA)

  • Zero-shot DA

  • One-shot DA

  • Few-shot DA

  • Source-free Unsupervised DA (SF-UDA)

  • Multi-source & Multi-target DA

  • Test-time Domain Adaptation

  • Black-box model adaptation

• Meta Learning / Learning to Learn

• Reliable Learning Across domains

  • Open-set DA

  • Partial-set DA

  • Universal DA

  • Anomaly Detection

  • Uncertainty Quantification in DA

  • Out-of-distribution Detection

  • Distribution Mismatch Measures

• Continual/Lifelong/Incremental Learning

• Multimodal learning

  • Cross-modal adaptation

  • Multimodal transfer learning

• Multimodal DA/DG

• Invariance and Equivariance in Deep Neural Networks

• Weakly-supervised Learning

• Self-supervised Learning

• Active Learning

• Federated Learning (FL)

  • Federated DA

  • Personalized FL

• Applications

  • Computer Vision

  • Robotics 

  • Autonomous Driving

  • Other domains

• Evaluation protocols

• Datasets

• ...

Important Dates

All deadlines are 11:59PM UTC-12:00 ("anywhere on Earth").

• Paper submissions are open https://cmt3.research.microsoft.com/ECMLPKDDworkshop2023/Track/42/Submission/Create

• Submission deadline: June 25th July 3rd, 2023 (extended)

• Notification of publication decision: July 12th, 2023

• Camera-ready due: August 4th, 2023

• Workshop: September 18th, 2023 (afternoon)

Best Paper Awards

The Program Committee and Organizers will select one (at least) of the accepted contributions for a best paper award.

Submission Instructions

Workshop papers must be prepared and submitted using the following Latex template: https://resource-cms.springernature.com/springer-cms/rest/v1/content/19238648/data/v6

This workshop allows for two different paper formats (maximum 10MB):

1. Short papers of maximum 4 pages (references included): not eligible for proceedings, cover early-stage ideas and/or work recently accepted, published or under review at other venues;

2. Full-length papers (maximum 8 pages, references included): eligible for proceedings (opt-in), cover novel work not previously published and not under review elsewhere.

A maximum of 2 optional supplementary material files (maximum 50MB) are allowed (Note: supplementary files can be uploaded from the Author Console only after the main manuscript has been uploaded):

• A PDF Appendix file using the same Latex tamplate as the main manuscript;

• A ZIP archive containing other supplementary material such as videos in MP4 format (maximum 3-minutes long) and/or code. A readme.txt file must be included in the archive to describe content.

The Workshops and Tutorials will be included in a joint Post-Workshop proceeding published by Springer Communications in Computer and Information Science, in 1-2 volumes, organised by focused scope and possibly indexed by WOS. Authors of full-length papers have the faculty to opt-in.
CCIS web page: https://www.springer.com/series/7899

We encourage authors who want to present and discuss their ongoing work to choose the Short Paper format or to opt-out from the proceedings, so that it can be potentially submitted concurrently to other venues and not count as a dual/double submission.

Recent work currently submitted to or accepted at other conferences and/or journals is still eligible for presentation at this workshop, but it is not eligible for inclusion in the proceedings.

Concurrent submissions to multiple ECML-PKDD 2023 Workshops are not allowed.

At least one registered author for each accepted paper has to take part in-person to the event in order to present the work.

For any inquiries or technical assistance, please contact the orgenizers via the form available at: https://sites.google.com/view/adapting-to-change-ecml-pkdd/contacts