Explaining complex systems: a tutorial on transparency and interpretability in machine learning models - Part II - Overview of global XAI methods

Aug 5, 2026

The tutorial’s second segment advances from local interpretability toward a broader survey of eXplainable AI (XAI) methods, extending the foundation established in Part I. Building on LIME and SHAP as representative local, model-agnostic explanation techniques, it covers global-oriented approaches including counterfactual explanations, equation discovery, and physics-informed AI integration. The segment contrasts with the earlier focus on two specific methods and provides a general understanding of explanation strategies for making complex machine learning models interpretable at a global scale.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

XAI transition from local to global

Global explanation techniques

Counterfactual explanations and equation discovery

Physics-informed AI integration

Introduction

Motivation for XAI in system identification

Local explanations in Part I (LIME, SHAP)

Shift to global explanation methods

Explaining complex systems Citation for published version (APA): Materassi, D. , Warnick, S. , Rojas, C. , Schoukens, M. , & Cross, E. (2024) . Explaining complex systems: a tutorial on transparency and interpretability in machine learning models (part II) . IFAC-PapersOnLine , 58(15), 497-501. [https://doi.org/10.1016/j.ifacol.2024.08.578](https://doi.org/10.1016/j.ifacol.2024.08.578) Document license: CC BY-NC-ND DOI: 10.1016/j.ifacol.2024.08.578 Document status and date: Published: 01/07/2024 Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication: • A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: [www.tue.nl/taverne](www.tue.nl/taverne) Take down policy If you believe that this document breaches copyright please contact us at: [openaccess@tue.nl](openaccess@tue.nl) providing details and we will investigate your claim. Download date: 13. Mar. 2025 [Available online at](Available online at www.sciencedirect.com)[ www.sciencedirect.com](Available online at www.sciencedirect.com) ScienceDirect IFAC PapersOnLine 58-15 (2024) 497–501 Explaining complex systems: a tutorial on transparency and interpretability in machine learning models (part II) Donatello Materassi ∗ Sean Warnick ∗∗ Cristian Rojas ∗∗∗ Maarten Schoukens ∗∗∗∗ Elizabeth Cross † ∗ Department of Electrical and Computer Engineering, University of Minnesota, [mater013@umn.edu](mater013@umn.edu) ∗∗ US Department of Homeland Security, Science and Technology Directorate, [sean.warnick@hq.dhs.gov](sean.warnick@hq.dhs.gov) ∗∗∗ EECS / Division of Decision and Control Systems, KTH Royal Institute of Technology, [cristian.rojas@ee.kth.se](cristian.rojas@ee.kth.se)[ ](cristian.rojas@ee.kth.se)∗∗∗∗ Control Systems Group, Eindhoven University of Technology, Eindhoven, The Netherlands, [m.schoukens@tue.nl](m.schoukens@tue.nl)[ ](m.schoukens@tue.nl)† Department of Mechanical Engineering, University of Sheffield, [e.j.cross@sheffield.ac.uk](e.j.cross@sheffield.ac.uk) Abstract: In the second segment of the tutorial, we transition from the granularity of local interpretability to a broader exploration of eXplainable AI (XAI) methods. Building on the specific focus of the first part, which delved into Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP), this section takes a more expansive approach. We will navigate through various XAI techniques of more global nature, covering counterfactual explanations, equation discovery, and th

FAQs

What does Part II of the tutorial focus on compared with Part I?

Part II shifts from local interpretability to a broader exploration of eXplainable AI (XAI) methods, emphasizing global explanation techniques.

Which XAI methods from Part I are referenced as the starting point for Part II?

Part II builds on Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) introduced in Part I.

What kinds of global explanation techniques are covered in Part II?

Part II covers counterfactual explanations, equation discovery, and the integration of physics-informed AI, providing an overview of these broader classes of techniques.

Aug 5, 2026

Upload by Theodore