Open AccessProceedings Article
A High Performance k-NN Classifier Using a Binary Correlation Matrix Memory
Ping Zhou,Jim Austin,John G. Kennedy +2 more
- 01 Dec 1998
- Vol. 11, pp 713-722
TL;DR: A novel and fast k-NN classifier that is based on a binary CMM (Correlation Matrix Memory) neural network, which gives over 200 times the speed of a current mid-range workstation, and is scaleable to very large problems.
read more
Abstract: This paper presents a novel and fast k-NN classifier that is based on a binary CMM (Correlation Matrix Memory) neural network. A robust encoding method is developed to meet CMM input requirements. A hardware implementation of the CMM is described, which gives over 200 times the speed of a current mid-range workstation, and is scaleable to very large problems. When tested on several benchmarks and compared with a simple k-NN method, the CMM classifier gave less than 1% lower accuracy and over 4 and 12 times speed-up in software and hardware respectively.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Emergent Neural Computational Architectures based on Neuroscience
Stefan Wermter,Jim Austin,David Willshaw +2 more
- 01 Jan 2001
TL;DR: The overall aim of the book is to present a broad spectrum of current research into biologically inspired computational systems and hence inspire the emergence of new computational approaches based on neuroscience to stimulate and encourage new research in this area.
109
A binary neural k -nearest neighbour technique
Victoria J. Hodge,Jim Austin +1 more
TL;DR: A novel k-NN classifier with linear growth and faster run-time built from binary neural networks is evaluated, demonstrating the superior performance with respect to speed and memory requirements of the binary approach compared to the standard approach.
2008 Special Issue: Neural network based pattern matching and spike detection tools and services - in the CARMEN neuroinformatics project
Martyn Fletcher,Bojian Liang,Leslie S. Smith,Alastair Knowles,Thomas Jackson,Mark Jessop,Jim Austin +6 more
TL;DR: The CARMEN project is described as well as the node infrastructure and an associated thick client tool for pattern visualisation and searching, the Signal Data Explorer (SDE), and new spike detection methods are discussed, which are central to the services provided by CARMEN.
24
Towards Novel Neuroscience-Inspired Computing
TL;DR: This overview incorporates some of the key research issues in the field of biologically inspired computing systems, such as modular organisation, robustness, timing and synchronisation, and learning and memory storage in the central nervous system.
15
A high performance k -NN approach using binary neural networks
TL;DR: This paper evaluates a novel k-nearest neighbour (k-NN) classifier built from binary neural networks and demonstrates the superior performance with respect to speed and memory requirements of the binary approach compared to the standard approach.
14
References
•Book
Machine Learning: Neural and Statistical Classification
Donald Michie,David Spiegelhalter,Charles C. Taylor,John A. Campbell +3 more
- 01 Jan 2009
TL;DR: A survey of previous comparisons and theoretical work descriptions of methods dataset descriptions criteria for comparison and methodology (including validation) empirical results machine learning on machine learning can be found in this article, where the authors also discuss their own work.
2.6K
Non-holographic associative memory
TL;DR: The features of a hologram that commend it as a model of associative memory can be improved on by other devices.
1K
Minimal consistent set (MCS) identification for optimal nearest neighbor decision systems design
Belur V. Dasarathy
- 01 Mar 1994
TL;DR: These results show that the nearest neighbor decision system performance suffers little degradation when the given large training set is replaced by its much smaller MCS in the operational phase of testing with an independent test set.
Fast implementations of nearest neighbor classifiers
TL;DR: Four techniques for expediting the nearest neighbor methods are described: replacing the linear search with a new kd tree method, exhibiting approximately O (N 1 2 ) behavior; employing an L ∞ instead of L 2 distance metric; using variance-ordered features; and rejecting prototypes by evaluating distances in low dimensionality subspaces.
81
Matching performance of binary correlation matrix memories
Mick Turner,Jim Austin +1 more
TL;DR: A theoretical framework for estimating the matching performance of binary correlation matrices acting as hetero-associative memories and the results highlight the fact that correlation-based models can act as highly efficient memories provided a small probability of retrieval error is accepted.
43
Related Papers (5)
Jim Austin
- 01 Feb 1998
Christopher M. Bishop
- 01 Jan 1995