## Algorithms for Lipschitz Learning on Graphs

published: Aug. 20, 2015, recorded: July 2015, views: 2925

# Slides

# Related content

# Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our**to describe your request and upload the data.**

__ticket system__*Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.*

# Description

We develop fast algorithms for solving regression problems on graphs where one is given the value of a function at some vertices, and must find its smoothest possible extension to all vertices. The extension we compute is the absolutely minimal Lipschitz extension, and is the limit for large $p$ of $p$-Laplacian regularization. We present an algorithm that computes a minimal Lipschitz extension in expected linear time, and an algorithm that computes an absolutely minimal Lipschitz extension in expected time $O (m n)$. The latter algorithm has variants that seem to run much faster in practice. These extensions are particulary amenable to regularization: we can perform $l_{0}$ regularization on the given values in polynomial time and $l_{1}$ regularization on the graph edge weights in time $\widetilde{O} (m^{3/2})$. Our algorithms naturally extend to directed graphs.

# Link this page

Would you like to put a link to this lecture on your homepage?

Go ahead! Copy the HTML snippet !

## Reviews and comments:

Larry Heiser, March 22, 2021 at 7:51 a.m.:Flame jets at the lower end heat all the materials in the kiln to high temperatures that range between 2,700 and 3,000 degrees Fahrenheit.

<a href="https://www.moosejawconcrete.com/">Concrete Elite</a>

Michael Keene, March 22, 2021 at 7:52 a.m.:This high heat drives off, or calcines, the chemically combined water and carbon dioxide from the raw materials and forms new compounds.

https://www.moosejawconcrete.com/

Karum, December 21, 2021 at 3:32 a.m.:I really learned a lot. Thanks.

www.hvaclancaster.com/

## Write your own review or comment: