Comparison of Some Methods for Processing 'Grey Level' Data in Weightless Networks

Mitchell, Richard; Box, Simon; Hawker, J.F.

doi:10.1142/9789812816849_0006

Cited by 18 publications

(11 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, in order to allow our WNN to deal with images, in which a pixel may assume a range of different values, we use minchinton cells [34]. Each neuron's synapse forms a minchinton cell with the next (the last one forms a minchinton cell with the first one).…”

Section: Handwritten Digit Recognition Systemmentioning

confidence: 99%

See 1 more Smart Citation

Fat-Fast VG-RAM WNN: A high performance approach

et al. 2016

View full text Add to dashboard Cite

This is the accepted version of the paper.This version of the publication may differ from the final published version. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Permanent repository link AbstractThe Virtual Generalizing Random Access Memory Weightless Neural Network (VG-RAM WNN) is a type of WNN that only requires storage capacity proportional to the training set. As such, it is an effective machine learning technique that offers simple implementation and fast training -it can be made in one shot. However, the VG-RAM WNN test time for applications that require many training samples can be large, since it increases with the size of the memory of each neuron. In this paper, we present Fat-Fast VG-RAM WNNs. Fat-Fast VG-RAM WNNs employ multi-index chained hashing for fast neuron memory search. Our chained hashing technique increases the VG-RAM memory consumption (fat) but reduces test time substantially (fast), while keeping most of its machine learning performance. To address the memory consumption problem, we employ a data clustering technique to reduce the overall size of the neurons' memory. This can be achieved by replacing clusters of neurons' memory by their respective centroid values. With our approach, we were able to reduce VG-RAM WNN test time and memory footprint, while maintaining a high and acceptable machine learning performance. We performed experiments with the Fat-Fast VG-RAM WNN applied to two recognition problems: (i) handwritten digit recognition, and (ii) traffic sign recognition. Our experimental results showed that, in both recognition problems, our new VG-RAM WNN approach was able to run three orders of magnitude faster and consume two orders of magnitude less memory than standard VG-RAM, while experiencing only a small reduction in recognition performance.

show abstract

Section: Handwritten Digit Recognition Systemmentioning

confidence: 99%

“…Instead, it is centered in the position of the Input Layer given by the inverse log-polar function of the neuron's position in the neural layer (see [24] for details). The VG-RAM synapses collect binary input vectors from the network's Input Layer using minchinton cells [34].…”

Section: Traffic Sign Recognition Systemmentioning

confidence: 99%

Fat-Fast VG-RAM WNN: A high performance approach

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Thus, in order to allow our VG-RAM WNN to deal with images, in which a pixel may assume a range of different values, we use minchinton cells (Mitchell et al, 1998). In the proposed VG-RAM WNN architectures, each neuron's synapse, w t , forms a minchinton cell with the next, w t+1 (w |W| forms a minchinton cell with w 1 ).…”

Section: Holistic Architecturementioning

confidence: 99%

VG-RAM Weightless Neural Networks for Face Recognition

Souza¹,

Badué²,

Pedroni³

et al. 2010

Face Recognition

View full text Add to dashboard Cite

“…Note that |X| < |V | (our experiments have shown that |X| < |V | provides better performance). Each neuron's synapse x i forms a minchinton cell with the next, x i+1 (x |X| forms a minchinton cell with x 1 ) [7]. The type of the minchinton cell we have used returns 1 if the synapse x i of the cell is connected to an input element n j whose value is larger than that of the element n k to which the synapse x i+1 is connected (i.e.…”

Section: Vg-ram Wnnmentioning

confidence: 99%