BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:"Machine learning in chemistry and beyond" (ChE-651) seminar by Ge
 emi Wellawatte: "Model Agnostic Counterfactual Explanations for Molecular 
 Property Predictions"
DTSTART:20221122T140000
DTEND:20221122T150000
DTSTAMP:20260510T075556Z
UID:075441373ddce41b0067d3d48f5f250e74a95a5dbbcd1c8639e02d31
CATEGORIES:Conferences - Seminars
DESCRIPTION:Geemi Wellawatte is a PhD student at the University of Rochest
 er\, NY\, with Prof. Andrew White. She completed her BSc. in Computational
  Chemistry in 2017 at the University of Colombo Sri Lanka. Her research wo
 rk focuses on developing computational models to solve chemistry problems\
 , particularly in coarse-grained modeling\, deep learning in chemistry\, a
 nd explainable AI.\nOne of the challenges with deep learning is lack of mo
 del interpretability. This is a significant drawback in the chemistry doma
 in\, as lack of knowledge "why a certain prediction was made" dissuades ch
 emists to trust predictions from deep learning. We propose a method that c
 an provide local explanations for arbitrary models with the use of molecul
 ar counterfactuals. These are sparse explanations composed of molecular st
 ructures. A counterfactual is an example as close to the original\, but wi
 th a different outcome. Although relatively new to AI\, counterfactual exp
 lanations are a mature topic in philosophy and mathematics. We use counter
 factuals to answer\, "what is the smallest change to the features that wou
 ld alter the prediction". Our Molecular Model Agnostic Counterfactual Expl
 anations (MMACE)\, method traverses a local chemical space around a given 
 base molecule to identify counterfactuals. Further\, we introduce an open-
 source software named “exmol” that implements the MMACE algorithm for 
 generating counterfactual explanations.
LOCATION:https://epfl.zoom.us/j/68447908297?pwd=OU5JUGJUSUhZc0ZNYjQ2WENvYl
 NRdz09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR