Distributions of Angles in Random Packing on Spheres

Abstract

This paper studies the asymptotic behaviors of the pairwise angles among n randomly and uniformly distributed unit vectors in R^p as the number of points n -> infinity, while the dimension p is either fixed or growing with n. For both settings, we derive the limiting empirical distribution of the random angles and the limiting distributions of the extreme angles. The results reveal interesting differences in the two settings and provide a precise characterization of the folklore that "all high-dimensional random vectors are almost always nearly orthogonal to each other". Applications to statistics and machine learning and connections with some open problems in physics and mathematics are also discussed.

Published in Journal of Machine Learning Research; 2013

References in corpus (3)

• The asymptotic distribution and Berry--Esseen bound of a new test for independence in high dimension with an application to stochastic optimization
• Some strong limit theorems for the largest entries of sample correlation matrices
• Surprising Asymptotic Conical Structure in Critical Sample Eigen-Directions

Cited by in corpus (35)

• Hopfield Networks is All You Need
• On high-dimensional sign tests
• Solving Systems of Random Quadratic Equations via Truncated Amplitude Flow
• Branches of triangulated origami near the unfolded state
• Are Discoveries Spurious? Distributions of Maximum Spurious Correlations and Their Applications
• Reshaped Wirtinger Flow and Incremental Algorithm for Solving Quadratic System of Equations
• Asymptotics of Empirical Eigen-structure for Ultra-high Dimensional Spiked Covariance Model
• Structured and Unstructured Outlier Identification for Robust PCA: A Non iterative, Parameter free Algorithm
• Surprising Asymptotic Conical Structure in Critical Sample Eigen-Directions
• Subspace clustering without knowing the number of clusters: A parameter free approach
• An overview of uniformity tests on the hypersphere
• Sparse Phase Retrieval via Truncated Amplitude Flow
• Spherical Cap Packing Asymptotics and Rank-Extreme Detection
• Any Orthonormal Basis in High Dimension is Uniformly Distributed over the Sphere
• Learning Rates as a Function of Batch Size: A Random Matrix Theory Approach to Neural Network Training
• Beyond Random Matrix Theory for Deep Networks
• A robust principal component analysis for outlier identification in messy microcalorimeter data
• Latent Variables on Spheres for Autoencoders in High Dimensions
• Guarding against Spurious Discoveries in High Dimensions
• Quantized Estimation of Gaussian Sequence Models in Euclidean Balls
• Uplink and Downlink Performance Bounds for Full Duplex Cellular Networks
• Norm-Explicit Quantization: Improving Vector Quantization for Maximum Inner Product Search
• Flatness is a False Friend
• Outperforming Word2Vec on Analogy Tasks with Random Projections
• Rank-Extreme Association of Gaussian Vectors and Low-Rank Detection
• Limiting behavior of largest entry of random tensor constructed by high-dimensional data
• Mean Test with Fewer Observation than Dimension and Ratio Unbiased Estimator for Correlation Matrix
• Statistical Query Lower Bounds for List-Decodable Linear Regression
• Fast, Parameter free Outlier Identification for Robust PCA
• Distributional Properties of Nearest-Site Angular Distances on the Sphere
• High-dimensional tests for spherical location and spiked covariance
• Angle distribution of two random chords in the disc: A sine law
• Circumspheres of sets of n+1 random points in the d-dimensional Euclidean unit ball (0<n<d+1)
• High-dimensional metric-measure limit of Stiefel and Grassmann manifolds
• On the Search for Tight Frames of Low Coherence