BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.0.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:326@isdm.umontpellier.fr
DTSTART;TZID=Europe/Paris:20260203T140000
DTEND;TZID=Europe/Paris:20260203T140000
DTSTAMP:20260828T082722Z
URL:https://isdm.umontpellier.fr/events/cross-validated-causal-inference-a
 -modern-method-to-combine-experimental-and-observational-data-2/
SUMMARY:Cross-Validated Causal Inference: a Modern Method to Combine Experi
 mental and Observational Data
DESCRIPTION:Online\n\nMachine Learning in Montpellier\, Theory &amp\; Pract
 ice - Xuelin Yang (UC Berkeley)\n\nWe develop new methods to integrate exp
 erimental and observational data in causal inference. While randomized con
 trolled trials offer strong internal validity\, they are often costly and 
 therefore limited in sample size. Observational data\, though cheaper and 
 often with larger sample sizes\, are prone to biases due to unmeasured con
 founders. To harness their complementary strengths\, we propose a systemat
 ic framework that formulates causal estimation as an empirical risk minimi
 zation (ERM) problem. A full model containing the causal parameter is obta
 ined by minimizing a weighted combination of experimental and observationa
 l losses—capturing the causal parameter’s validity and the full model
 ’s fit\, respectively. The weight is chosen through crossvalidation on t
 he causal parameter across experimental folds. Our experiments on real and
  synthetic data show the efficacy and reliability of our method. We also p
 rovide theoretical non-asymptotic error bounds.\n\nMachine Learning in Mon
 tpellier\, Theory &amp\; Practice
ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/
 2026/06/ml-mtp-gC78d5.png
CATEGORIES:ML MTP
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20251026T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR