posted by user: biagiolarosa || 2352 views || tracked by 3 users: [display]

XAI 4 Debugging 2021 : eXplainable AI approaches for debugging and diagnosis


When Dec 13, 2021 - Dec 14, 2021
Where NeurIPS2021
Submission Deadline Sep 15, 2021
Notification Due Oct 10, 2021
Final Version Due Oct 20, 2021
Categories    explainable ai   machine learning   visual analytics   XAI

Call For Papers

The workshop aims at collecting novel methods and discussing challenges, issues, and goals around the usage of XAI approaches to debug, understand and improve current deep learning models. In particular, we aim to bring together researchers from two communities that share the same goal on the topic: the eXplainable artificial intelligence and the visual analytics communities.

=== Topics ===

- Novel XAI methods (post-hoc, ante-hoc, model-agnostic, etc.);
- Interpretable/Explainable deep learning models;
- Applications and protocols that use current XAI methods to improve and/or debug deep learning models;
- XAI evaluation: how can we assess the quality of explanations and their effectiveness for debugging purposes?
- Visualization techniques for debugging Deep Learning models;
- Debugging via interpretability: How can explainable artificial intelligence help us in debugging deep learning models?
- Methods to identify and address sources and causes of failure ( data, regularization, objective functions, etc.);
- Visual analytics systems for understanding and debugging deep learning models;
- Visual analytics systems guided by XAI methods - where XAI methods are the core of the system;
- Analysis of limitations of current approaches;
- Position papers on the topic of the workshop.

=== Tracks and Important dates ===
Regular Track
Submissions to the main track of the workshop have to be full papers, position papers, and papers describing open problems on one of the topics listed above. They must be novel contributions of various lengths. Accepted papers will be presented as contributed talks during the workshop, or during a poster section. Extended versions of selected papers will be considered for publication in a journal special issue.

September 15, 2021 – Submission deadline.
October 15, 2021 – Notification date

Mentorship track
This track aims at helping young researchers to polish their papers before submission to other venues. It is directed to the researchers that don't have access to resource and mentors. Papers submitted to this track won't be presented during the workshop, but they will receive feedback from our mentors.
September 30, 2021 – Submission deadline.

A glimpse of the future track
This is a special track for MSc and 1st year Ph.D. students.
Submissions must include research plans and the agenda of young researchers covering one of the topics of the workshop. The goal is to introduce them into the community, giving them the possibility to present their plan during a workshop talk.

September 20, 2021 – Submission deadline.
October 10, 2021 – Notification date

== Organizers ==
Roberto Capobianco, Sony AI & Sapienza University of Rome
Biagio La Rosa, Sapienza University of Rome
Leilani Gilpin, Sony AI
Wen Sun, Cornell University
Alice Xiang, Sony AI
Alexander Feldman, PARC

Related Resources

EAIH 2024   Explainable AI for Health
Ei/Scopus-AACIP 2024   2024 2nd Asia Conference on Algorithms, Computing and Image Processing (AACIP 2024)-EI Compendex
xAI 2024   The 2nd World Conference on eXplainable Artificial Intelligence
ICMLA 2024   23rd International Conference on Machine Learning and Applications
xAI ST Actionable XAI 2024   xAI World Conference Special Track on Actionable Explainable AI
IEEE-Ei/Scopus-SGGEA 2024   2024 Asia Conference on Smart Grid, Green Energy and Applications (SGGEA 2024) -EI Compendex
FLAIRS-37 ST XAI, Fairness, and Trust 2024   FLAIRS-37 Special Track on Explainable, Fair, and Trustworthy AI
ICDM 2024   IEEE International Conference on Data Mining
EXTRAAMAS 2024   EXplainable and TRAnsparent AI and Multi-Agent Systems
AIMLAI@ECML-PKDD 2024   AIMLAI@ECML/PKDD 2024: Joint Tutorial on XAI for Sequential Data and International Workshop on Advances in Interpretable Machine Learning and Artificial Intelligence