BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250822T115804Z
LOCATION:Room 6.0D13
DTSTART;TZID=Europe/Stockholm:20250617T153000
DTEND;TZID=Europe/Stockholm:20250617T160000
UID:submissions.pasc-conference.org_PASC25_sess107_msa234@linklings.com
SUMMARY:Anomaly Detection with a Deep Abstaining Classifier Model Under Fe
 derated Learning
DESCRIPTION:Cristina Garcia Cardona and Jamal Mohd-Yusof (Los Alamos Natio
 nal Laboratory)\n\nA deep abstaining classifier, or DAC, introduced first 
 for combating label noise, is a regular deep neural network classifier (DN
 N) but with an additional (abstain) class and a custom loss function that 
 permits abstention during training. This allows the DAC to identify and ab
 stain on (or decline to classify) confusing samples, without the need for 
 manually labeling these cases, while continuing to learn and improve class
 ification performance on the non-abstained samples. It has been shown that
  the resulting models can significantly improve accuracy compared to the o
 riginal DNN, at the cost of reduced coverage. The DAC learns patterns in t
 he data that make prediction unreliable, is more robust to feature noise, 
 and constitutes a useful tool to diagnose uncertain predictions at inferen
 ce time. In this talk, we describe how we adapt the DAC to be combined wit
 h Federated Learning (FL) to allow for a distributed training using data f
 rom different silos that cannot be openly shared due to privacy concerns, 
 as is frequently encountered when dealing with health records. We also dem
 onstrate how this DAC+FL model can be applied for anomaly detection and di
 scuss how to configure this per silo.\n\nDomain: Applied Social Sciences a
 nd Humanities, Life Sciences, Computational Methods and Applied Mathematic
 s\n\nSession Chair: Adam Spannaus (Oak Ridge National Laboratory)\n\n
END:VEVENT
END:VCALENDAR