AN INTEGER PROGRAMMING MODEL FOR DESIGNING CAUSAL DISCOVERY NETWORKS by Ali \u0130lhan Halilo\u011flu<\/p>\n
Thesis Advisor: Prof. Dr. Oya Kara\u015fan
\nCo-Advisor: Assoc. Prof. \u00d6zlem Karsu<\/p>\n
Date: August 07 2024, Wednesday
\nTime: 15:30
\nPlace: Zoom<\/p>\n
This is an online seminar. To obtain event details please send a message to department.<\/p>\n
Abstract:<\/p>\n
We propose a novel mixed integer programming formulation for the design of causal discovery networks. The model takes a set of rules that indicate statistical dependency relations between features of a given dataset, the so-called d-connection and d-separation relations, and aims to fit a casual network with minimum (weighted) violations. Allowing feedback cycles and latent confounders, our formulation stands out from most of the existing attempts in the literature. Although our model can work as an unsupervised machine learning model, it possesses the necessary flexibility for the decision-maker to enter known causal relations. The performance of our model is tested with several synthetic datasets.<\/p>\n","protected":false},"excerpt":{"rendered":"
AN INTEGER PROGRAMMING MODEL FOR DESIGNING CAUSAL DISCOVERY NETWORKS by Ali \u0130lhan Halilo\u011flu Thesis Advisor: Prof. Dr. Oya Kara\u015fan Co-Advisor: Assoc. Prof. \u00d6zlem Karsu Date: August 07 2024, Wednesday Time: 15:30 Place: Zoom This is an online seminar. To obtain event details please send a message to department. Abstract: We [\u2026] <\/p>\n