BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:CAM Colloquium - Nikita Doikov\, Department of Operations Rese
 arch & Information Engineering\, Cornell University
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260518T154549Z
UID:tag:localist.com\,2008:EventInstance_52276965551632
DTSTART:20260313T194500Z
DTEND:20260313T204500Z
DESCRIPTION:Title: Gradient-Normalized Smoothness for Optimization with App
 roximate Hessians\n\nAbstract: In this talk\, we discuss the design of opt
 imization algorithms that use approximate second-order information to achi
 eve fast global convergence rates for both convex and non-convex objective
 s. The key innovation of our analysis is a novel notion called Gradient-No
 rmalized Smoothness\, which characterizes the maximum radius of a ball aro
 und the current point that yields a good relative approximation of the gra
 dient field. Our theory establishes a natural intrinsic connection between
  Hessian approximation and the linearization of the gradient. Importantly\
 , Gradient-Normalized Smoothness does not depend on the specific problem c
 lass of the objective functions\, while effectively translating local info
 rmation about the gradient field and Hessian approximation into the global
  behavior of the method.  This new concept equips approximate second-order
  algorithms with universal global convergence guarantees\, recovering stat
 e-of-the-art rates for functions with Hölder-continuous Hessians and thir
 d derivatives\, quasi-self-concordant functions\, as well as smooth classe
 s in first-order optimization. These rates are achieved automatically and 
 extend to broader classes\, such as generalized self-concordant functions.
  We demonstrate direct applications of our results for global linear rates
  in logistic regression and softmax problems with approximate Hessians\, a
 s well as in non-convex optimization using Fisher and Gauss-Newton approxi
 mations.\n\nBio: Nikita Doikov joined Cornell in January 2026 as an Assist
 ant Professor in the School of Operations Research and Information Enginee
 ring. He received his PhD from UCLouvain\, Belgium\, under the supervision
  of Yurii Nesterov in 2021. After that\, he was a postdoctoral researcher 
 at EPFL\, Switzerland\, working in the Machine Learning and Optimization L
 aboratory led by Martin Jaggi. He also gained industrial experience as an 
 intern at Google in 2016 and 2018.
GEO:42.443451;-76.481506
LOCATION:Frank H. T. Rhodes Hall\, 655
SUMMARY:CAM Colloquium - Nikita Doikov\, Department of Operations Research 
 & Information Engineering\, Cornell University
URL;VALUE=URI:https://events.cornell.edu/event/cam-colloquium-nikita-doikov
 -department-of-operations-research-information-engineering-cornell-univers
 ity
CATEGORIES:Colloquium
END:VEVENT
END:VCALENDAR