Fast Cross-Polytope Locality-Sensitive Hashing
February 22, 2016 Β· Declared Dead Β· π Information Technology Convergence and Services
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Christopher Kennedy, Rachel Ward
arXiv ID
1602.06922
Category
cs.DS: Data Structures & Algorithms
Cross-listed
cs.CG
Citations
14
Venue
Information Technology Convergence and Services
Last Checked
3 months ago
Abstract
We provide a variant of cross-polytope locality sensitive hashing with respect to angular distance which is provably optimal in asymptotic sensitivity and enjoys $\mathcal{O}(d \ln d )$ hash computation time. Building on a recent result (by Andoni, Indyk, Laarhoven, Razenshteyn, Schmidt, 2015), we show that optimal asymptotic sensitivity for cross-polytope LSH is retained even when the dense Gaussian matrix is replaced by a fast Johnson-Lindenstrauss transform followed by discrete pseudo-rotation, reducing the hash computation time from $\mathcal{O}(d^2)$ to $\mathcal{O}(d \ln d )$. Moreover, our scheme achieves the optimal rate of convergence for sensitivity. By incorporating a low-randomness Johnson-Lindenstrauss transform, our scheme can be modified to require only $\mathcal{O}(\ln^9(d))$ random bits
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Data Structures & Algorithms
π
π
The Cartographer
R.I.P.
π»
Ghosted
Route Planning in Transportation Networks
R.I.P.
π»
Ghosted
Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration
R.I.P.
π»
Ghosted
Hierarchical Clustering: Objective Functions and Algorithms
R.I.P.
π»
Ghosted
Graph Isomorphism in Quasipolynomial Time
π
π
The Cartographer
Simulation optimization: A review of algorithms and applications
Died the same way β π» Ghosted
R.I.P.
π»
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
π»
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
π»
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
π»
Ghosted