BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Cambridge Ellis Unit Seminar Series - Dr Cheng Zhang-  9 February 
 2022- 2pm - Dr Cheng Zhang
DTSTART:20220209T140000Z
DTEND:20220209T150000Z
UID:TALK169559@talks.cam.ac.uk
CONTACT:Kimberly Cole
DESCRIPTION:Causal inference is essential for data-driven decision-making 
 across domains such as business engagement\, medical treatment or policy-m
 aking.  However\, research on causal discovery and inference has evolved s
 eparately\, and the combination of the two domains is nontrivial.  In this
  talk\, I will present our Deep End-to-end Causal Inference (DECI) framewo
 rk\, a single flow-based method that takes in observational data and can p
 erform both causal discovery and inference\, including conditional average
  treatment effect estimation (CATE). We provide a theoretical guarantee th
 at DECI can recover the ground truth under mild assumptions. In addition\,
  our method can handle heterogeneous\, real-world\, mixed-type data with m
 issing values\, allowing for both continuous and discrete treatment decisi
 ons. Moreover\, the design principle of our method can generalize beyond D
 ECI\, providing a general End-to-end Causal Inference (ECI) recipe\, which
  enables different ECI frameworks to be built using existing methods. Our 
 results show the superior performance of DECI when compared to relevant ba
 selines for both causal discovery and (C)ATE estimation on over a thousand
  experiments\, with both synthetic datasets and various other causal machi
 ne learning benchmark datasets. We hope that our work bridges the causal d
 iscovery and inference communities.\n 
LOCATION:https://eng-cam.zoom.us/j/86919451784?pwd=N2JLSWdhWUs1U3JEVTZVY0J
 QWmM2QT09 
END:VEVENT
END:VCALENDAR