Parallel Approximate Undirected Shortest Paths Via Low Hop Emulators
Offered By: Paul G. Allen School via YouTube
Course Description
Overview
Explore parallel approximate undirected shortest paths algorithms in this theory seminar presented by Peilin Zhong from Columbia University. Delve into the concept of low hop emulators and their application in solving graph problems efficiently. Learn about the Bellman-Ford algorithm, hopset construction, and the "Law of Amateur" in graph theory. Discover applications in submillimetre and strong submillimetre problems, and understand the highway example used to illustrate key concepts. Analyze the construction and performance of these algorithms, and consider open problems in the field. This comprehensive lecture, recorded on February 25, 2020, includes closed captions and is part of the Paul G. Allen School's theory seminar series.
Syllabus
Introduction
Problem Statement
Bellman for Organ
Hopset
Original Construction
Main Question
Results
Law of Amateur
Local Amateur
Applications
Submillimetre
Strong Submillimetre
Highway Example
Construction
Analysis
Open Problems
Taught by
Paul G. Allen School
Related Courses
Automata TheoryStanford University via edX Intro to Theoretical Computer Science
Udacity Computing: Art, Magic, Science
ETH Zurich via edX 理论计算机科学基础 | Introduction to Theoretical Computer Science
Peking University via edX Quantitative Formal Modeling and Worst-Case Performance Analysis
EIT Digital via Coursera