BEGIN:VCALENDAR
PRODID:-//github.com/ical-org/ical.net//NONSGML ical.net 4.0//EN
VERSION:2.0
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:STANDARD
DTSTART:20241103T020000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250309T020000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
DESCRIPTION:The convergence of foundation models and agentic AI is transfo
 rming how we approach personal health. In this talk\, I present our work 
 across three interconnected directions. First\, I describe our efforts bu
 ilding personal health agents — from early LLM-based coaching systems to 
 multi-agent architectures that combine data analysis\, domain reasoning\,
  and behavioral coaching. Second\, I discuss wearable foundation models t
 hat learn general representations from large-scale sensor data\, enabling
  zero and few-shot health understanding across diverse tasks. Third\, I p
 resent recent work on the science of scaling agent systems\, where we der
 ive quantitative scaling principles governing the interplay between agent
  quantity\, coordination structure\, and model capability. We identify ke
 y effects including tool-coordination trade-offs\, capability saturation\
 , and topology-dependent error amplification that inform how to design re
 liable multi-agent health systems. Together\, these threads point toward 
 a future where AI systems can natively understand personal health data an
 d deliver actionable guidance to everyone. Xin Liu\, PhD \, is a Senior R
 esearch Scientist at Google Health Research\, where he leads research at 
 the intersection of artificial intelligence\, wearable sensing\, and pers
 onal health. He received his PhD in Computer Science from the University 
 of Washington in 2023. Before UW\, he obtained his bachelor's degree from
  the University of Massachusetts Amherst. His research spans personal hea
 lth agents\, wearable foundation models\, and large language model reason
 ing\, with multiple systems deployed in Google products reaching billions
  of users. His work has been published in venues including Nature Medicin
 e\, Nature Communications\, Science Advances\, NeurIPS\, ICLR\, ACL\, and
  EMNLP\, accumulating over 4000 citations. He has received the Google PhD
  Fellowship\, the ACM IMWUT Distinguished Paper Award\, multiple NeurIPS 
 and ICLR Oral highlights. He is also a core contributor to Google's Gemin
 i 2.5 models. His long-term vision is to build intelligent systems that m
 ake personalized health accessible to everyone.\n\nSpeaker:\nXin Liu\, Ph
 D\n\nAdmission:\nFree\n\nFood:\nLunch\n\nDetails URL:\nhttps://medicine.y
 ale.edu/event/research-in-progress-or-rising-star-seminar-xinliu/\n
DTEND;TZID=America/New_York:20260212T130000
DTSTAMP:20260501T122149Z
DTSTART;TZID=America/New_York:20260212T120000
LOCATION:ZOOM \, URL: https://yale.zoom.us/j/97037559157?from=addon
SEQUENCE:0
STATUS:Confirmed
SUMMARY:Research in Progress | Rising Star Seminar
UID:0badab63-85df-4bba-a1f3-cb0231c72e70
END:VEVENT
END:VCALENDAR