BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Hands-On eXplainable Artificial Intelligence (XAi)
DTSTART;VALUE=DATE-TIME:20231112T060000Z
DTEND;VALUE=DATE-TIME:20231112T133000Z
DTSTAMP;VALUE=DATE-TIME:20261009T022400Z
UID:indico-event-177@events.kfupm.edu.sa
DESCRIPTION:Speakers: Faizan Ahmed (University of Twente)\, Adeel Ahmad (C
 OMSATS University Islamabad)\n\nThis hands-on workshop introduces particip
 ants to eXplainable Artificial Intelligence (XAi)\, focusing on techniques
  that enhance the interpretability of AI models. Participants will gain th
 eoretical knowledge and practical experience to effectively apply XAi meth
 ods.Requirements:Basic understanding of programming conceptsLaptop with Py
 thon and relevant libraries (scikit-learn\, LIME\, SHAP) installedSample d
 atasets and code templates will be providedLearning Outcomes: By the end o
 f the workshop\, participants will:Understand the basics of machine learni
 ng and its typesGrasp the fundamentals of interpretable machine learningBe
  able to apply LIME\, SHAP\, intrinsic XAi\, global XAi\, and feature impo
 rtance techniques to explain AI model predictionsHave hands-on experience 
 building and interpreting interpretable models\, assessing feature importa
 nce\, and understanding model behaviorAgenda:Session 1: Introduction to Ma
 chine Learning (1.5 hour)Session 2: Fundamentals of Interpretable Machine 
 Learning (1 hour)Session 3: Intrinsic and Global XAi Methods (1.5 hour)Ses
 sion 4: Local XAi Methods (1.5 hours)Session 5: Conclusion and Certificate
  Distribution (30 minutes)Brief overview of machine learning concepts and 
 types (supervised\, unsupervised\, reinforcement)Understanding training\, 
 validation\, and testing datasetsIntroduction to evaluation metrics (accur
 acy\, precision\, recall\, F1-score)Practical Exercise: Dara Driven Physic
 al Modelling Explaining interpretable machine learning vs. black-box mode
 lsCommonly used interpretable models: linear regression\, decision trees\,
  etc.Pros and cons of interpretable modelsExploring inherently interpretab
 le models\, such as decision trees and linear modelsHow intrinsic methods 
 contribute to model transparency?Permutation feature importance: Understan
 ding how shuffling feature values impacts model performanceModel-specific 
 methods for feature importance: Exploring coefficients in linear models an
 d impurity reduction in decision treesIntroduction to LIME and its princip
 lesAnalyzing LIME results and understanding local explanationsUnderstandin
 g SHAP values and their significancePractical exercise: Implementing LIME 
 to explain a black-box model's predictions OR Applying SHAP to explain a c
 omplex model's output Interpreting SHAP plots and summary plotsSummarizin
 g key takeaways from the workshopCertificate DistributionCertification:  
             The participant will receive a certificate after comple
 tion of this workshop\n\nhttps://events.kfupm.edu.sa/event/177/
LOCATION:60/1-Auditorium (Administration Building)
URL:https://events.kfupm.edu.sa/event/177/
END:VEVENT
END:VCALENDAR