Special Seminar given by Sofia Triantafyllou, PhD, Postdoctoral Researcher, Department of Biomedical Engineering, University of Pennsylvania

You are invited to a Special Seminar on Friday, September 15th at 11:00 AM given by Sofia Triantafyllou, PhD, Postdoctoral Researcher, Department of Biomedical Engineering, University of Pennsylvania.

The seminar is titled “Logic-based causal discovery” and will be held at the Department of Biomedical Informatics (address below).

Abstract:

Causal modeling allows predicting a system’s behavior not only under observation but also under intervention. Computational causal discovery reverse-engineers causal models (networks) from observational data with limited or no interventions.  In this work, I will present logic-based causal discovery, a new, versatile approach for learning causal networks from observations and interventions:  based on standard causal assumptions, associative patterns in the data that constrain the search space of possible causal models are expressed as a logic formula. Truth-setting assignments to this formula correspond to causal networks that fit the data. This approach can reason with multiple data sets, handle conflicting statistical information, and produce novel, non-trivial predictions. I will also discuss possible applications and future extensions, aiming to answer specific causal questions in biomedical informatics.

 

LOCATION: Department of Biomedical Informatics
5607 Baum Blvd., Fourth Floor, Conference Room 407 A/B
Pittsburgh, PA 15206

 

Questions:
Linda Mignogna412-648-9254/Lkm16@pitt.edu
Date of Event: 
Friday, September 15, 2017 - 11:00
Post Date: 
Thursday, August 10, 2017
^