DiscoverMicrosoft Research (Audio) - Channel 9Pacific Northwest Probability Seminar: Gravitational Allocation to Uniform Points on the Sphere
Pacific Northwest Probability Seminar: Gravitational Allocation to Uniform Points on the Sphere

Pacific Northwest Probability Seminar: Gravitational Allocation to Uniform Points on the Sphere

Update: 2017-11-20
Share

Description

Given n uniform points on the surface of a two-dimensional sphere, how can we partition the sphere fairly among them? "Fairly" means that each region has the same area. It turns out that if the given points apply a two-dimensional gravity force to the rest of the sphere, then the basins of attraction for the resulting gradient flow yield such a partition - with exactly equal areas, no matter how the points are distributed. (See the cover of the AMS Notices at http://www.ams.org/publications/journals/notices/201705/rnoti-cvr1.pdf.) Our main result is that this partition minimizes, up to a bounded factor, the average distance between points in the same cell. I will also present an application to almost optimal matching of n uniform blue points to n uniform red points on the sphere, connecting to a classical result of Ajtai, Komlos and Tusnady (Combinatorica 1984). Joint work with Nina Holden and Alex Zhai. 

Comments 
In Channel
loading
00:00
00:00
x

0.5x

0.8x

1.0x

1.25x

1.5x

2.0x

3.0x

Sleep Timer

Off

End of Episode

5 Minutes

10 Minutes

15 Minutes

30 Minutes

45 Minutes

60 Minutes

120 Minutes

Pacific Northwest Probability Seminar: Gravitational Allocation to Uniform Points on the Sphere

Pacific Northwest Probability Seminar: Gravitational Allocation to Uniform Points on the Sphere

MSRVideo