Office of Scientific and International Cooperation

سخنرانی به زبان انگليسي با موضوع:

برآورد سازگار اثرات شبکه ای پنهان برمبنای مشاهدات کرانه ای

ميزبان: دكتر سعيد رضاخواه، استاد دانشكده رياضي و علوم كامپيوتر، دانشگاه صنعتي اميركبير

سخنران: Prof. Claudia Klüppelberg، استاد دانشگاه صنعتي مونيخ، آلمان

چهارشنبه 18 اسفندماه 1400، ساعت 20:30

Abstract: This talk contributes to the recently emerged topic of graphical models for extremes, in particular to extreme value models on directed graphs. In this context, the Latent River Problem has emerged as a flagship problem for causal discovery in extreme value statistics. We provide a simple and efficient algorithm QTree, to solve the Latent River Problem that outperforms existing methods. QTree returns a directed graph and achieves almost perfect recovery on the Upper Danube, the existing benchmark dataset, as well as on new data from the Lower Colorado River in Texas. It can handle missing data and has an automated parameter tuning procedure. In our paper, we also show that, under a Bayesian network model for extreme values with propagating noise, the QTree algorithm returns for $n\to\infty$ a.s. the correct tree. This is joint work with Ngoc M. Tran and Johannes Buck.