BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Approximation Polynomial-Time Algorithms for Minimum Cost Interven
 tion in Causal Effect Identification
DTSTART:20230829T140000
DTEND:20230829T160000
DTSTAMP:20260407T123355Z
UID:66c46354139bdedcfbc242bd0390801f61f2c4321e7f28a7ac781915
CATEGORIES:Conferences - Seminars
DESCRIPTION:Sepehr Elahi\nEDIC candidacy exam\nExam president: Prof. Micha
 el Gastpar\nThesis advisor: Prof. Patrick Thiran\nThesis co-advisor: Prof.
  Negar Kiyavash\nCo-examiner: Prof. Ola Svensson\n\nAbstract\nWe investiga
 te two fundamental problems in causality: causal discovery and causal iden
 tifiability. For the former\, we summarize the important findings of two w
 orks\, the first of which focuses on learning a causal graph through small
 -sized interventions\, and the second of which looks at the problem of lea
 rning a causal graph through single-target noisy interventions. For the la
 tter\, we discuss a work that formulated the problem of ensuring identifia
 bility through minimum cost interventions as a combinatorial optimization 
 problem. We also briefly present our own work on the problem of minimum co
 st intervention for causal effect identification\, where we derive bounds 
 on the performance of a polynomial-time algorithm on random graphs.\n\nBac
 kground papers\n\n	Learning Causal Graphs with Small Interventions (2015)
 \n	Sample Efficient Active Learning of Causal Trees (2019)\n	Minimum Cost
  Intervention Design for Causal Effect Identification (2022)\n
LOCATION:BC 229 https://plan.epfl.ch/?room==BC%20229
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR