September 2025

An Overview to

Concept-based Explainable AI

ECML-PKDD 2025, Porto, Portugal

A tutorial to dive into the world of

Concept-based Explainable AI

Introduction

This tutorial provides a comprehensive introduction to Concept-Based Explainable AI (C-XAI), a novel approach within the broader field of eXplainable Artificial Intelligence (XAI) that enhances AI interpretability by shifting from traditional feature-based explanations to high-level, human-understandable concepts.

This tutorial aims to deliver an in-depth analysis of the fundamental principles and methodologies associated with C-XAI. The tutorial will comprise five different sections, each presented by a different instructor. Specifically, it will present:

An overview of traditional XAI methods and their limitations, emphasizing the need for concept-based explainability
A rigorous introduction to the C-XAI domain, including what a concept and a concept-based explanation are, along with a taxonomy of C-XAI methods covering thirteen dimensions
An in-depth analysis of the two principal paradigms within C-XAI, namely post-hoc and explainable by design, detailing both foundational and cutting-edge approaches
Guidance on the selection of the most appropriate C-XAI approach, as well as discussing the challenges of current C-XAI methods and future directions
A hands-on session to develop practical skills, where participants will directly implement some of the presented methodologies.

Key Details

The following are key details and dates for this tutorial:

Tutorial Date: TBD
Conference: ECML PKDD 2025 - The 2025 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases.
Location: Porto, Portugal.
Modality: In-person half-day event.

Submit your aper

Attending the Tutorial

The tutorial is organized as a part of the 2025 ECML PKDD conference that will be held in Porto, Portugal from Monday 15th until Friday 19th of September 2025. To attend the tutorial, you will need a single day registration for the day of the tutorial (TBD) or a full conference registration. Please look at registration details, including costs, on the official registration site.

We recommend potential attendees to register early to the conference as (1) the “early bird” registration price ends soon, and (2), if you need a Schengen visa, it is recommended to apply for this visa as early as possible to guarantee the visa’s arrival before the workshop start date.

Program

Registration Open 8:30 - 9:15

9:15 - 10:00

Introduction to XAI techniques

Elena Baralis

10:00 - 10:30

Taxonomy of C-XAI

Eliana Pastor

Break 10:30 - 11:00

11:00 - 11:45

Post-hoc vs By-design C-XAI approaches

Gabriele Ciravegna

11:45 - 12:15

Guidance for selecting C-XAI approaches

Tania Cerquitelli

12:15 - 13:15

Hands-on session

Eleonora Poeta

Launch Break 1:00 - 2:00

Speakers

Eleonora Poeta

Eleonora Poeta is a PhD student in Explainable and Trustworthy Artificial Intelligence at the Politecnico di Torino, Italy. She earned her Master's degree in Data Science and Engineering from the same institution in April 2023. Her research focuses on Trustworthy AI, with particular interests in explainable AI (XAI), concept-based explainability, and robustness in AI. She is one of the organizers of the Special Track on Concept-based Explainable AI at the 3rd World Conference on eXplainable Artificial Intelligence 2025.

Gabriele Ciravegna

Gabriele Ciravegna is a Postdoctoral Researcher at Politecnico di Torino, focusing on advancing Artificial Intelligence. His research aims at enhancing the comprehensibility, reliability, and robustness of neural networks. Since 2019, he has been publishing and reviewing for top conferences and journals like Neurips, ICML, and IEEE TPAMI. For his contributions to Explainable AI, he has been invited to speak at several venues. Gabriele completed his Ph.D. under Prof. Marco Gori, earning the IEEE Caianiello Award for Best Ph.D. Thesis in 2023. Since 2022, he has been teaching Machine Learning at Politecnico di Torino and Université Côte d'Azur.

Eliana Pastor

Eliana Pastor is an assistant professor at Politecnico di Torino, Italy. She received her PhD in Computer and Control Engineering from Politecnico di Torino in October 2021. Her doctoral thesis, “Pattern-based Algorithms for Explainable AI”', addresses the lack of interpretability of machine learning algorithms for the perspective of individual predictions and data subgroups. She is the lecturer of the `Explainable and Trustworthy AI' course at Politecnico di Torino. Her research interests are trustworthy AI, explainable AI, and fairness in AI. She co-organized the full-day workshop XKDD: Workshop on eXplainable Knowledge Discovery in Data Mining at ECML PKDD 2024.

Tania Cerquitelli

Tania Cerquitelli is a full professor at the Department of Control and Computer Engineering of the Politecnico di Torino, Italy. Her research activities have been mainly devoted to fostering and sharing research and innovation on data science and machine learning, with a particular focus on explainable AI. Tania has designed and developed novel, scalable, and general-purpose algorithms to extract new forms of knowledge patterns in different application domains, from Industry 4.0 to energy-related applications. She co-organized the full-day workshop XKDD: Workshop on eXplainable Knowledge Discovery in Data Mining at ECML PKDD 2023.

Elena Baralis

Elena Baralis is a full professor at the Department of Control and Computer Engineering of the Politecnico di Torino since January 2005. She is currently the Deputy Rector of the Politecnico di Torino. Her current research interests are in the field of machine learning and data science, more specifically on explainable AI, bias detection in data analytics, and machine learning algorithms for big data. She has published over 200 papers in international journals and conference proceedings. She was general co-chair of the 30th ECML PKDD in 2023 and the program co-chair of IEEE ICDM 2024. She has served as area chair and in the program committees of international conferences and workshops, among which IEEE ICDM, ACM SIGMOD, VLDB, ACM SAC, ACM CIKM, ECML PKDD.

Page updated

Google Sites

Report abuse