Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[theory-seminar] Fwd: ISL Colloquium on Thursday Oct. 10: Sam Hopkins (UC Berkeley), "Algorithms for robust and heavy-tailed statistics -- theory and experiment", 4:30-5:30pm, Packard 101

Kabir Chandrasekher kabirc at stanford.edu
Mon Oct 7 07:40:22 PDT 2019


Hi All,

This week, Sam Hopkins, will be giving an ISL colloquium that may be of
interest.  The details are below.

Thanks,
Kabir

---------- Forwarded message ---------
From: Kabir Chandrasekher <kabirc at stanford.edu>
Date: Mon, Oct 7, 2019 at 7:39 AM
Subject: ISL Colloquium on Thursday Oct. 10: Sam Hopkins (UC Berkeley),
"Algorithms for robust and heavy-tailed statistics -- theory and
experiment", 4:30-5:30pm, Packard 101
To: <isl-colloq at lists.stanford.edu>, <
information_theory_forum at lists.stanford.edu>, <
ee-students-forum at lists.stanford.edu>, <ee-postdocs at lists.stanford.edu>, <
cs-students-announce at lists.stanford.edu>
Cc: Sam Hopkins <hopkins at berkeley.edu>


*Title: *Algorithms for robust and heavy-tailed statistics -- theory and
experiment

*Speaker: *Sam Hopkins (Berkeley)

*Time & location:* Thursday October 10, 4:30-5:30 pm, Packard 101
Coffee and pastries will be served before the talk at 4pm in the Packard
second floor kitchen

*Abstract:* Algorithms for statistics on corrupted or heavy-tailed data
have seen a flurry activity in the last few years. (Indeed, even the last
few months!) I will survey some recent developments, and then zoom in on
joint work with Yihe Dong and Jerry Li in which we focus on translating the
progress in polynomial-time algorithms into something practical. In
particular, we obtain the first nearly-linear time algorithm for robust
mean estimation in high dimensions, where the goal is to estimate the mean
of a random vector from independent samples of which a constant fraction
have been maliciously corrupted. Our algorithm is sufficiently practical
that our implementation scales to thousands of dimensions and tens/hundreds
of thousands of samples on laptop hardware; I will discuss some
experimental validations of our theoretical results.

Based on "Quantum Entropy Scoring for Fast Robust Mean Estimation and
Improved Outlier Detection," to appear in NeurIPS 2019.
https://arxiv.org/pdf/1906.11366.pdf
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/theory-seminar/attachments/20191007/b7e10bf7/attachment.html>


More information about the theory-seminar mailing list