BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Reliable machine learning in the wild
DTSTART:20220217T160000
DTEND:20220217T170000
DTSTAMP:20260407T102921Z
UID:9b4aca1d6081cd37acadd019722dfbb4e0bf5b3e54dba0fd56f48cff
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Pang Wei Koh - Stanford University\nIC Faculty candidate\n
 \nAbstract\nMachine learning systems are widely deployed today\, but they 
 are unreliable. They can fail – and with catastrophic consequences – o
 n subpopulations of the data\, such as particular demographic groups\, or 
 when deployed in different environments from what they were trained on. In
  this talk\, I will describe our work towards building reliable machine le
 arning systems that are robust to these failures. First\, I will show how 
 we can use influence functions to understand the predictions and failures 
 of existing models through the lens of their training data. Second\, I wil
 l discuss the use of distributionally robust optimization to train models 
 that perform well across all subpopulations. Third\, I will describe WILDS
  – a benchmark of in-the-wild distribution shifts spanning applications 
 such as pathology\, conservation\, remote sensing\, and drug discovery –
  and show how current state-of-the-art methods\, which perform well on syn
 thetic distribution shifts\, still fail to be robust on these real-world s
 hifts. Finally\, I will describe our work on building more reliable COVID-
 19 models\, using anonymized cellphone mobility data\, to inform public he
 alth policy\; this is a challenging application as the underlying environm
 ent is often changing and there is substantial heterogeneity across demogr
 aphic subpopulations.\n\nBio\nPang Wei Koh is a PhD student at Stanford\, 
 advised by Percy Liang. He studies the theory and practice of building rel
 iable machine learning systems. His research has been published in Nature 
 and Cell\, featured in media outlets such as The New York Times and The Wa
 shington Post\, and recognized by a Meta Research PhD fellowship\, best pa
 per awards at ICML and KDD\, and the Kennedy Prize for best honors thesis 
 at Stanford. Prior to his PhD\, he was the 3rd employee and Director of Pa
 rtnerships at Coursera. \n\nMore information
LOCATION:https://epfl.zoom.us/j/65850986919?pwd=bVhITWZFalVRRmphYWFITGJmck
 w2QT09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR