BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260601T203852EDT-9472CDxC4D@132.216.98.100
DTSTAMP:20260602T003852Z
DESCRIPTION:Justin Weltz\, PhD\n\nIs an Emerging Political Economies and Ap
 plied Complexity Postdoctoral Fellow at the Santa Fe Institute\, Duke Univ
 ersity\n\nNOTE: Meet & Greet Justin Weltz from 3-3:30pm in Room 1140\n\nWH
 EN: Wednesday\, October 8\, 2025\, from 3:30 to 4:30 p.m.\n	WHERE: Hybrid |
  2001 McGill College Avenue\, Rm 1140\; Zoom\n	NOTE: Justin Weltz will be p
 resenting in-person at SPGH\n\nAbstract\n\nRespondent-driven sampling (RDS
 ) is widely used to study hidden or hard-to-reach populations by incentivi
 zing study participants to recruit their social connections. The success a
 nd efficiency of RDS can depend critically on the nature of the incentives
 \, including their number\, value\, call to action\, etc. Standard RDS use
 s an incentive structure that is set a priori and held fixed throughout th
 e study. Thus\, it does not make use of accumulating information on which 
 incentives are effective and for whom. We propose a reinforcement learning
  (RL) based adaptive RDS study design in which the incentives are tailored
  over time to maximize cumulative utility during the study. We show that t
 hese designs are more efficient\, cost-effective\, and can generate new in
 sights into the social structure of hidden populations. In addition\, we d
 evelop methods for valid post-study inference which are non-trivial due to
  the adaptive sampling induced by RL as well as the complex dependencies a
 mong subjects due to latent (unobserved) social network structure. We prov
 ide asymptotic regret bounds and illustrate its finite sample behavior thr
 ough a suite of simulation experiments.\n\nSpeaker Bio\n\nI am an Emerging
  Political Economies and Applied Complexity Postdoctoral Fellow at the San
 ta Fe Institute\, where I work with Matt Jackson\, Eleanor Power\, and Fio
 na Steele on statistical inference for complex network sampling techniques
 . I recently completed my Ph.D. at Duke University advised by Alexander Vo
 lfovsky and Eric Laber. My dissertation research focused on creating metho
 ds for the study and assistance of hard-to-reach populations.\n\nWebsite: 
 justinweltz.com\n\n \n
DTSTART:20251008T193000Z
DTEND:20251008T203000Z
SUMMARY:Reinforcement Learning for Respondent-Driven Sampling
URL:https://www.mcgill.ca/epi-biostat-occh/channels/event/reinforcement-lea
 rning-respondent-driven-sampling-368109
END:VEVENT
END:VCALENDAR