BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Non-rational Infinite-Horizon Control: A Unified Framework for Rob
 ust Control through Frequency-Domain Optimization
DTSTART:20241113T160000
DTEND:20241113T170000
DTSTAMP:20260414T203213Z
UID:86b23ee284ddfaebdce34982ee004cf741b7967107d9e5aa4e95e384
CATEGORIES:Conferences - Seminars
DESCRIPTION:Taylan Kargin\, PhD candidate in Electrical Engineering\, Calt
 ech\, USA\nABSTRACT: \nInfinite-horizon controllers provide scalable alte
 rnatives to finite-horizon designs with long-term performance and stabilit
 y guarantees. However\, synthesizing them is challenging due to their infi
 nite-dimensional nature and the non-triviality of causality and other info
 rmation constraints. In this talk\, I will introduce a novel framework for
  scalable and efficient synthesis of a broad class of infinite-horizon con
 trol problems. This framework unifies several classical designs—such as 
 H2/LQG\, H∞\, and regret-optimal control—and extends to advanced setti
 ngs\, including risk-sensitive\, Wasserstein distributionally robust\, Hp-
 norm\, and mixed-criteria designs. By leveraging convex duality\, the fram
 ework reformulates the original problem into a convex dual optimization ov
 er positive-definite operators\, simplifying the incorporation of informat
 ion constraints. It also enables efficient frequency-domain computation vi
 a variations of first-order methods such as projected gradient ascent and 
 Frank-Wolfe. While the resulting controllers may have non-rational transfe
 r functions\, I will present a novel minimax rational approximation techni
 que that yields near-optimal and stabilizing state-space controllers of an
 y desired degree. I will conclude with numerical experiments demonstrating
  the framework’s effectiveness for distributional robust and mixed-crite
 ria control.\n\nBIOGRAPHY:\nTaylan Kargin is a PhD candidate in Electrical
  Engineering at Caltech\, advised by Prof. Babak Hassibi. His research lie
 s at the intersection of control theory\, optimization\, and machine learn
 ing\, with a focus on robust and adaptive control of dynamical systems und
 er uncertainty. He earned his M.S. in Electrical Engineering from Caltech 
 in 2023 and dual bachelor’s degrees in Electrical and Electronics Engine
 ering and Physics from Bilkent University in 2019. As an undergraduate\, h
 e researched at the National Magnetic Resonance Research Center under the 
 guidance of Prof. Tolga Çukur. Taylan also gained industry experience as 
 an intern at Amazon\, where he collaborated with Dr. Kevin Small on reinfo
 rcement learning for large language models.
LOCATION:https://epfl.zoom.us/j/66099708879
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR