BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The interpretability wars - Rachel C. Zhang (DAMTP PhD)
DTSTART:20260204T110000Z
DTEND:20260204T120000Z
UID:TALK243859@talks.cam.ac.uk
CONTACT:Rachel Zhang
DESCRIPTION:As AI systems become increasingly deployed in science and high
 -risk domains\, interpretability has emerged as a critical concern. But wh
 at does "interpretability" actually mean\, and is it even achievable? This
  journal club examines five recent papers + articles that offer different 
 answers to these questions. \n\nThe journal club will discuss:\n\nBarbiero
  et al.\, 2025: Foundations of Interpretable Models https://arxiv.org/pdf/
 2508.00545\n\nRowan et al.\, 2025: On the Definition and Importance of Int
 erpretability in Scientific Machine Learning https://arxiv.org/pdf/2505.13
 510\n\nMeloux et al.\, 2025: The Dead Salmons of AI Interpretability https
 ://arxiv.org/pdf/2512.18792\n\nRudin 2019: Stop Explaining Black Box Machi
 ne Learning Models for High Stakes Decisions and Use\nInterpretable Models
  Instead https://arxiv.org/pdf/1811.10154\n\nAnd an article by Hendrycks +
  Hiscott in 2025: The Misguided Quest for Mechanistic AI Interpretability 
 https://ai-frontiers.org/articles/the-misguided-quest-for-mechanistic-ai-i
 nterpretability\n
LOCATION:MR10\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR