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:20250822T115809Z
LOCATION:Room 6.0D13
DTSTART;TZID=Europe/Stockholm:20250617T150000
DTEND;TZID=Europe/Stockholm:20250617T170000
UID:submissions.pasc-conference.org_PASC25_sess107@linklings.com
SUMMARY:MS4D - Biopreparadness at Scale via Context-Aware Agent-Based Mode
 ls
DESCRIPTION:A rapid response to the initial phase of the COVID-19 pandemic
  was hampered by decentralized data collection, analysis, and the novelty 
 of the virus itself; vital metrics for virus characteristics, such as its 
 transmissibility and virulence, were unknown. Moreover, the disease progre
 ssion was spatially heterogeneous; different regions experienced waves at 
 varying times and with differing intensities. To mitigate these challenges
  and better inform public health officials for the next pandemic, we are d
 eveloping methods to assimilate real-world data into biologically informed
  agent-based models, facilitating biopreparedness at scale in near-real ti
 me. These models will allow for population stratification along multiple c
 omorbidities or socio demographic factors across diverse geospatial region
 s by incorporating decentralized data in a mathematically private way. By 
 incorporating this data from varied populations across a region, these mod
 els will assist public health agencies in mitigating an emerging outbreak 
 and effectively managing hospital capacity. We will highlight different co
 mputational methods designed to address these key bioprepardness challenge
 s in this minisymposium.\n\nAnomaly Detection with a Deep Abstaining Class
 ifier Model Under Federated Learning\n\nA deep abstaining classifier, or D
 AC, introduced first for combating label noise, is a regular deep neural n
 etwork classifier (DNN) but with an additional (abstain) class and a custo
 m loss function that permits abstention during training. This allows the D
 AC to identify and abstain on (or decline to...\n\n\nCristina Garcia Cardo
 na and Jamal Mohd-Yusof (Los Alamos National Laboratory)\n----------------
 -----\nData-Driven Agent Based Modeling for Precision Public Health\n\nHig
 h-quality, accurate, and real-time information about disease spread is cri
 tical for rapid response to a biothreat. Electronic health data is collect
 ed by approximately 90% of all physicians in the United States. However, i
 ts broad use for public health surveillance and monitoring is inhibited by
  ...\n\n\nHeidi Hanson (Oak Ridge National Laboratory)\n------------------
 ---\nSpeeding Up LLM Inference via Sequential Speculative Decoding\n\nAs L
 arge Language Models (LLMs) grow in size and capability, their high comput
 ational cost poses a major challenge for real-time applications, making ef
 ficient inference a critical research problem. Speculative Decoding (SD) h
 as emerged as a promising technique to accelerate LLM inference by leverag
 ...\n\n\nMeiyu Zhong, Noel Teku, and Ravi Tandon (The University of Arizon
 a)\n---------------------\nEnhanced Uncertainty Quantification in Air Poll
 ution Models and Impact on Epidemiological Risk\n\nAdvancements in remote 
 sensing, geospatial data, physicochemical and source apportionment models,
  citizen science networks and machine learning have greatly improved our a
 bility to predict air pollution at high spatiotemporal resolution and over
  large domains and time periods. Air pollution models w...\n\n\nRima Habre
  (University of Southern California)\n\nDomain: Applied Social Sciences an
 d Humanities, Life Sciences, Computational Methods and Applied Mathematics
 \n\nSession Chair: Adam Spannaus (Oak Ridge National Laboratory)
END:VEVENT
END:VCALENDAR