BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:XAI 2025
DTSTART:20250626T070000Z
DTEND:20250626T120000Z
DTSTAMP:20260608T224900Z
UID:indico-event-5116@agenda.ciemat.es
CONTACT:miguel.cardenas@ciemat.es\;pablo.garcia@ciemat.es
DESCRIPTION:Hands-on course in Explainable Artificial IntelligenceWe are p
 leased to announce that a Hands-on Course in Explainable Artificial Intell
 igence (XAI) will take place in the context of the 14th Iberian Gravitatio
 nal Waves Meeting (IGWM2025)\, hosted at  CIEMAT (Madrid). Participation 
 is open to IGWM2025 attendees\, as well as to members of REDONGRA or IPARC
 OS\, regardless of whether they attend the conference.DescriptionThis prac
 tical course will equip participants with foundational knowledge and hands
 -on experience in Explainable Artificial Intelligence (XAI) techniques\, f
 ocusing on their application to gravitational wave data analysis and relat
 ed scientific fields. Through practical examples\, participants will explo
 re how XAI algorithms identify and highlight the most relevant features of
  input data that contribute to model predictions\, enhancing the interpret
 ability of AI models in physics by clarifying the relationship between inp
 ut signals and model outputs. The course is especially well-suited for res
 earchers and students interested in improving model transparency in scient
 ific applications. All code and materials used during the session will be 
 provided.The course will take place in Room B on the second floor of Build
 ing 1. For practical information\, please refer to the IGWM2025 web page.C
 ontentsGrad-CAM Algorithm — Rationale\, Strengths\, and WeaknessesExplor
 e the application of the Grad-CAM algorithm in a classification task using
  a simplified version of the Gravity Spy dataset. LIME Algorithm — Rati
 onale\, Strengths\, and WeaknessesExplore how the LIME algorithm operates 
 in a similar classification context\, utilising a reduced Gravity Spy data
 set. SHAP Algorithm — Rationale\, Strengths\, and WeaknessesExamine the
  SHAP algorithm in action\, applied to a classification task involving a s
 ubset of galaxies from the COSMOS2015 catalogue. DatasetsThe selected dat
 asets were chosen based on several key criteria: their small size\, facili
 tating hands-on demonstrations\; the interpretability of their outputs\; a
 nd their generality\, which enhances the transferability of insights to ot
 her domains. Required Resources and PrerequisitesParticipants will need a
  laptop with internet access (eduroam connectivity will be provided by CIE
 MAT) and a valid Google account to access Google Drive and Google Colabora
 tory. Prior knowledge of neural networks\, including experience with Tenso
 rFlow\, is required\, as all code examples are implemented using this fram
 ework. Familiarity with Python libraries such as NumPy\, Pandas\, and Matp
 lotlib is also assumed.RegistrationThis course is open exclusively to IGWM
 2025 attendees\, as well as to members of REDONGRA or IPARCOS\, regardless
  of whether they attend the conference. Participation is free of charge\; 
 however\, registration is mandatory for all attendees. The deadline to reg
 ister is 20 June.To register\, please follow this link or click on Registr
 ation in the menu on the left.\n\nhttps://agenda.ciemat.es/event/5116/
IMAGE;VALUE=URI:https://agenda.ciemat.es/event/5116/logo-2904877615.png
LOCATION:Room B (CIEMAT (Madrid\, Spain))
URL:https://agenda.ciemat.es/event/5116/
END:VEVENT
END:VCALENDAR