DSC/e Lecture Alexander Nikolaev, Mutual Information Matching (MIM) for Fast Observational Causal Inference

Date:
22 August
Time:
12:30 - 13:30
Location:
Auditorium 14, Den Dolech 2, 5612 AZ Eindhoven
Subscription:
From 08 August
Organizer:
DSC/e
Subscribe for this event

Data mining techniques are good at discovering frequent patterns and predictive models or correlations. Data mining aiming at prescriptive analytics, i.e. causal knowledge discovery, is challenging – causal relationships are typically identified from randomized (control) experiments while data mining typically deals with observational data. The primary approach used over the past thirty years to find the subset of data that closely resembles experimental data has been to individually match treated and untreated (control) units. Dr. Nikolaev from the University at Buffalo, NY will present a novel optimization-based approach to construct such datasets automatically. The technical part of the talk is based on his recent JMLR paper: Link

Abstract Alexander Nikolaev
Bio Alexander Nikolaev

PROGRAM

Date and time: Tuesday, August 22, 12:30 – 13.30
Location: Auditorium, room AUD 14, Den Dolech 2, 5612 AZ Eindhoven

12:00-12:30 Welcome
12.30-13.30 Lecture by Assistant Professor Alexander Nikolaev

The DSC/e is organizing several interesting events. If you are interested in a certain lecture or workshop please send an email to dsce@remove-this.tue.nl or subscribe for our newsletter here.