BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Foundation models for personal health signals - Dimitris Spathis\,
  Google
DTSTART:20260305T140000Z
DTEND:20260305T150000Z
UID:TALK235897@talks.cam.ac.uk
CONTACT:Cecilia Mascolo
DESCRIPTION:The unprecedented success of foundation models has transformed
  our understanding of artificial intelligence\, yet their application to p
 ersonal health and ubiquitous sensing remains a complex frontier. In this 
 talk\, I will share my journey in building AI for health monitoring\, star
 ting with early efforts to improve data efficiency\, robustness\, and fair
 ness through self-supervision. A key milestone in this journey was the dev
 elopment of PaPaGei\, the first open foundation model for photoplethysmogr
 aphy (PPG). While PaPaGei demonstrated the utility of pre-training on vari
 ous biosignal datasets\, the rigorous demands of real-world applications\,
  such as accurate activity recognition\, heart rate\, or other biomarker m
 onitoring in dynamic environments\, highlight the need for even greater ge
 neralization and scale. These challenges motivate the transition to Large 
 Sensor Models that address these hurdles by scaling up both model size and
  the diversity of user data. This scaling unlocks emergent benefits that s
 maller models cannot achieve\, positioning such foundation models to becom
 e the backbone of any future sensing task.\n\nBio: Dimitris Spathis is a r
 esearch scientist at Google and a visiting researcher at the University of
  Cambridge\, where he completed his PhD. He was previously a senior resear
 ch scientist at Nokia Bell Labs\, leading efforts in AI for multimodal hea
 lth. During his studies\, he worked at Microsoft Research\, Telefonica\, a
 nd Ocado\, while in 2020 he helped start one of the largest studies in aud
 io AI for health (covid-19-sounds.org). He is particularly interested in w
 ays that foundation models and data from personal devices can be helpful f
 or daily health. He studies various topics in AI including data-efficiency
 \, multimodality\, model robustness/fairness\, and signal processing. His 
 recent work has been featured in international media outlets such as the N
 ew York Times\, BBC\, CNN\, Guardian\, Washington Post\, Forbes\, and Fina
 ncial Times. He serves on the program committees of top AI conferences suc
 h as AAAI\, IJCAI\, and KDD\, as well as the editorial boards of Nature Di
 gital Medicine & IEEE Pervasive Computing. More details: https://dispathis
 .com/
LOCATION:Computer Lab\, LT2 and Online
END:VEVENT
END:VCALENDAR