BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/1641
DTSTAMP:20251120T093528Z
SUMMARY:Sparse Mean Estimation in Adversarial Settings via Incremental Lear
 ning
DESCRIPTION:Speaker: Santanu Das (TIFR)\n\nAbstract: \nSparse mean estimati
 on is a fundamental problem in high-dimensional statistics\, arising in di
 verse applications such as signal processing\, genomics\, and machine lear
 ning. However\, real-world datasets are rarely clean—samples are often c
 orrupted by adversarial noise or malicious outliers. This motivates the st
 udy of robust sparse mean estimation\, where the goal is to design estimat
 ors that remain accurate even when a fraction of the data has been arbitra
 rily contaminated.\nIn this talk\, we discuss the recent paper “Sparse M
 ean Estimation in Adversarial Settings via Incremental Learning”\, which
  provides a new perspective on achieving robustness through Hadamard param
 eterization. While Hadamard parameterization has proven useful in classica
 l sparse estimation tasks\, this paper demonstrates how it can be leverage
 d to obtain a provably robust sparse mean estimation algorithm. The method
  combines the structural benefits of Hadamard parameterisation with previo
 usly known robust estimation techniques. The resulting estimator achieves 
 strong performance guarantees in adversarial settings while maintaining co
 mputational efficiency.\n
URL:https://www.tcs.tifr.res.in/web/events/1641
DTSTART;TZID=Asia/Kolkata:20251121T160000
DTEND;TZID=Asia/Kolkata:20251121T170000
LOCATION:A-201 (STCS Seminar Room)
END:VEVENT
END:VCALENDAR