Explainable Artificial Intelligence
Programma
Course Introduction - Motivation for XAI- What is an explanation - The role of “Explanation” in the novel ML process: the assessment guidelines for trustworthy (2h)
The taxonomy of XAI methods for Machine Learning- Overview of explanation methods (2)
Overview of post-hoc explanation methods and evaluation metrics (2h)
Post hoc methods (on Images data) - IntGrad, Lime, GRADCAM(++), SmoothGRAD (2h)
Post hoc methods (on Images data) – Post hoc methods using latent space (examplars and counterexamplars) - ABELE (2h)
Contrastative Reasoning: counterfactual a causality (2h)
Hands-on: on XAI methods from XAI Library – Post hoc methods (tabular) (Lore Lime Anchor Shap) (4h)
Overview of transparent by-design explanation methods (2h)
Overview of transparent by-design explanation methods - Explaining by design & Global Explainer: EBM, Tlace, TREPAN, GlocalX, Protonet, Interpretable Models (2h)
Hands-on: on Post hoc methods (Images) – project presentation - IntGrad, Lime, GRADCAM(++), SmoothGRAD, (4h)
Interactive XAI - seminar (2h)
Explaining by design with Knowledge Injection - seminar (2h)
Hands-on: global methods and project (2h)
Obiettivi formativi
Black box AI systems for automated decision-making, often based on machine learning over (big) data, map a user’s features into a class or a score without exposing the reasons why. This is problematic not only for the lack of transparency but also for possible biases inherited by the algorithms from human prejudices and collection artifacts hidden in the training data, which may lead to unfair or wrong decisions. The future of AI lies in enabling people to collaborate with machines to solve complex problems. Like any efficient collaboration, this requires good communication, trust, clarity, and understanding. Explainable AI addresses such challenges and for years different AI communities have studied such topics, leading to different definitions, evaluation protocols, motivations, and results. This (30 hours) course provides a reasoned introduction to the work of Explainable AI (XAI) to date, and surveys the literature with a focus on machine learning and symbolic AI-related approaches. We motivate the needs of XAI in real-world and large-scale applications, while presenting state-of-the-art techniques and best practices, as well as discussing the many open challenges. An XAI platform with a collection of many of the recently proposed algorithms will be presented on specific use cases and it will be possible to familiarize with some of the methods.
Moduli
| Modulo | Ore | CFU | Docenti |
|---|---|---|---|
| Didattica integrativa | 20 | 0 | Gizem Gezici, Roberto Pellungrini |
| Explainable Artificial Intelligence | 20 | 3 | Fosca Giannotti |