Fast k-NNG construction with GPU-based quick multi-select.

PLoS One

Department of Mechanical Engineering, Complex Systems Simulation Lab, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin, United States of America.

Published: October 2015

In this paper, we describe a new brute force algorithm for building the k-Nearest Neighbor Graph (k-NNG). The k-NNG algorithm has many applications in areas such as machine learning, bio-informatics, and clustering analysis. While there are very efficient algorithms for data of low dimensions, for high dimensional data the brute force search is the best algorithm. There are two main parts to the algorithm: the first part is finding the distances between the input vectors, which may be formulated as a matrix multiplication problem; the second is the selection of the k-NNs for each of the query vectors. For the second part, we describe a novel graphics processing unit (GPU)-based multi-select algorithm based on quick sort. Our optimization makes clever use of warp voting functions available on the latest GPUs along with user-controlled cache. Benchmarks show significant improvement over state-of-the-art implementations of the k-NN search on GPUs.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4014471PMC
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0092409PLOS

Publication Analysis

Top Keywords

brute force
8
algorithm
5
fast k-nng
4
k-nng construction
4
construction gpu-based
4
gpu-based quick
4
quick multi-select
4
multi-select paper
4
paper describe
4
describe brute
4

Similar Publications

A CRISPR-Cas and Argonaute-Driven Two-Factor Authentication Strategy for Information Security.

ACS Nano

January 2025

Department of Clinical Laboratory of Sir Run Run Shaw Hospital, College of Biosystems Engineering and Food Science, Zhejiang University School of Medicine, Hangzhou 310058, People's Republic of China.

The escalating growth in computing power and the advent of quantum computing present a critical threat to the security of modern cryptography. Two-factor authentication strategies can effectively resist brute-force attacks to improve the security of access control. Herein, we proposed a two-factor and two-authentication entity strategy based on the trans-cleavage activity of CRISPR-Cas and the "dual-step" sequence-specific cleavage of Argonaute.

View Article and Find Full Text PDF

Image Encryption Method Based on Three-Dimensional Chaotic Systems and V-Shaped Scrambling.

Entropy (Basel)

January 2025

School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China.

With the increasing importance of securing images during network transmission, this paper introduces a novel image encryption algorithm that integrates a 3D chaotic system with V-shaped scrambling techniques. The proposed method begins by constructing a unique 3D chaotic system to generate chaotic sequences for encryption. These sequences determine a random starting point for V-shaped scrambling, which facilitates the transformation of image pixels into quaternary numbers.

View Article and Find Full Text PDF

Exploring the conformational space of molecules remains a challenge of fundamental importance to quantum chemistry: identification of relevant conformers at ambient conditions enables predictive simulations of almost arbitrary properties. Here, we propose a novel approach, called TTConf, to enable conformational sampling of large organic molecules where the combinatorial explosion of possible conformers prevents the use of a brute-force systematic conformer search. We employ tensor trains as a highly efficient dimensionality reduction algorithm, effectively reducing the scaling from exponential to polynomial.

View Article and Find Full Text PDF

Algebraic structures are highly effective in designing symmetric key cryptosystems; however, if the key space is not sufficiently large, such systems become vulnerable to brute-force attacks. To address this challenge, our research focuses on enlarging the key space in symmetric key schemes by integrating the non-chain ring with a four-dimensional chaotic system. While chaotic maps offer significant potential for data processing, relying solely on them does not fully leverage their operational advantages.

View Article and Find Full Text PDF

Intelligent two-phase dual authentication framework for Internet of Medical Things.

Sci Rep

January 2025

Department of Computer Science, College of Computer and Information Sciences, King Saud University, Riyadh, 11543, Saudi Arabia.

The Internet of Medical Things (IoMT) has revolutionized healthcare by bringing real-time monitoring and data-driven treatments. Nevertheless, the security of communication between IoMT devices and servers remains a huge problem because of the inherent sensitivity of the health data and susceptibility to cyber threats. Current security solutions, including simple password-based authentication and standard Public Key Infrastructure (PKI) approaches, typically do not achieve an appropriate balance between security and low computational overhead, resulting in the possibility of performance bottlenecks and increased vulnerability to attacks.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!