Quasiperiodic-Chaotic Neural Networks and Short-Term Analog Memory

2021 ◽  
Vol 31 (01) ◽  
pp. 2130003
Author(s):  
Natsuhiro Ichinose
Keyword(s):  
Neural Networks ◽  
Visual Memory ◽  
Large Scale ◽  
Quasiperiodic Solution ◽  
Short Term ◽  
Chaotic Neural Networks ◽  
Analog Memory ◽  
Short Term Visual Memory ◽  
Potential Applications ◽  
Chaotic Neuron

A model of quasiperiodic-chaotic neural networks is proposed on the basis of chaotic neural networks. A quasiperiodic-chaotic neuron exhibits quasiperiodic dynamics that an original chaotic neuron does not have. Quasiperiodic and chaotic solutions are exclusively isolated in the parameter space. The chaotic domain can be identified by the presence of a folding structure of an invariant closed curve. Using the property that the influence of perturbation is conserved in the quasiperiodic solution, we demonstrate short-term visual memory in which real numbers are acceptable for representing colors. The quasiperiodic solution is sensitive to dynamical noise when images are restored. However, the quasiperiodic synchronization among neurons can reduce the influence of noise. Short-term analog memory using quasiperiodicity is important in that it can directly store analog quantities. The quasiperiodic-chaotic neural networks are shown to work as large-scale analog storage arrays. This type of analog memory has potential applications to analog computation such as deep learning.

A Feature-Segmentation Model of Short-Term Visual Memory

Perception ◽  
10.1068/p3320 ◽  
2002 ◽  
Vol 31 (5) ◽  
pp. 579-589 ◽  
Author(s):  
Koji Sakai ◽  
Toshio Inui
Keyword(s):  
Visual Memory ◽  
Random Variable ◽  
Short Term ◽  
Pattern Complexity ◽  
Convex Part ◽  
Good Account ◽  
Feature Segmentation ◽  
Short Term Visual Memory ◽  
The Relationship ◽  
Response Feature

A feature-segmentation model of short-term visual memory (STVM) for contours is proposed. Memory of the first stimulus is maintained until the second stimulus is observed. Three processes interact to determine the relationship between stimulus and response: feature encoding, memory, and decision. Basic assumptions of the model are twofold: (i) the STVM system divides a contour into convex parts at regions of concavity; and (ii) the value of each convex part represented in STVM is an independent Gaussian random variable. Simulation showed that the five-parameter fits give a good account of the effects of the four experimental variables. The model provides evidence that: (i) contours are successfully encoded within 0.5 s exposure, regardless of pattern complexity; (ii) memory noise increases as a linear function of retention interval; (iii) the capacity of STVM, defined by pattern complexity (the degree that a pattern can be handled for several seconds with little loss), is about 4 convex parts; and (iv) the confusability contributing to the decision process is a primary factor in deteriorating recognition of complex figures. It is concluded that visually presented patterns can be retained in STVM with considerable precision for prolonged periods of time, though some loss of precision is inevitable.

Short Term Visual Memory and the Pulfrich Phenomenon

Nature ◽  
1969 ◽  
Vol 222 (5194) ◽  
pp. 639-641 ◽  
Author(s):  
BELA JULESZ ◽  
BENJAMIN WHITE
Keyword(s):  
Visual Memory ◽  
Short Term ◽  
Short Term Visual Memory

Capacity Limitation in Short-Term Visual Memory for Contour Curvature

Perception ◽  
10.1068/p3365 ◽  
2005 ◽  
Vol 34 (9) ◽  
pp. 1095-1105 ◽  
Author(s):  
Koji Sakai
Keyword(s):  
Size Effect ◽  
Visual Memory ◽  
Weber Fraction ◽  
Short Exposure ◽  
Difference Threshold ◽  
Short Term ◽  
Capacity Limitation ◽  
Set Size ◽  
Short Term Visual Memory ◽  
Contour Curvature

I measured the difference threshold for contour curvature in short-term visual memory (STVM) using a two-interval forced-choice partial discrimination task. In experiments 1 and 2, the study stimulus consisting of 1 to 4 curved contours was briefly presented. It was followed by a single contour stimulus after a retention interval. The subjects judged if the test stimulus had a higher or lower curvature than the corresponding study contour. The results of experiment 1 showed that the Weber fraction increased monotonically with increasing set size. The results of experiment 2 clarified that the set-size effect was not due to a temporal limitation in encoding resulting from the short exposure time. In experiment 3, the study stimuli always consisted of 4 items, but the numbers of memorised items were different in each condition. Nevertheless, the results showed the set-size effect, which indicated that its occurrence depended largely on the capacity limitation in short-term visual memory (STVM) storage. Otherwise, the Weber fraction was not hugely higher for set size 4 compared with set size 1. It was concluded that only 1 object could be retained in STVM with high fidelity, but that at least 4 objects could be retained in STVM with some degree of fidelity.

scholarly journals Serial position effects in short-term visual memory: A SIMPLE explanation?

2007 ◽  
Vol 35 (1) ◽  
pp. 176-190 ◽  
Author(s):  
Dennis C. Hay ◽  
Mary M. Smyth ◽  
Graham J. Hitch ◽  
Neil J. Horton
Keyword(s):  
Serial Position ◽  
Visual Memory ◽  
Simple Explanation ◽  
Short Term ◽  
Position Effects ◽  
Serial Position Effects ◽  
Short Term Visual Memory

scholarly journals The storage of physical and name properties in short-term visual memory

1976 ◽  
Vol 4 (1) ◽  
pp. 6-10 ◽  
Author(s):  
Robert A. Reeve ◽  
Ralph Hall
Keyword(s):  
Visual Memory ◽  
Short Term ◽  
Short Term Visual Memory

New Estimate for Storage Time in Sensory Memory

1984 ◽  
Vol 59 (3) ◽  
pp. 683-686 ◽  
Author(s):  
Jesse E. Purdy ◽  
Kelly M. Olmstead
Keyword(s):  
Visual Memory ◽  
Interstimulus Interval ◽  
Storage Time ◽  
Recognition Task ◽  
Sensory Memory ◽  
Short Term ◽  
Short Term Visual Memory ◽  
Visual Storage ◽  
Sensory Storage ◽  
Information Transfers

Sperling in 1960 reported information in sensory storage remained for about one sec. In 1974 Phillips reported that information in sensory storage passed on to short-term visual memory after 100 msec. To distinguish between these alternatives, 55 subjects received 36 trials in which two matrices of letters, familiar shapes, or non-familiar shapes were presented successively in a recognition task. The interstimulus interval varied systematically. Results showed that as the interval increased, performance decreased. Further, memory for letters and familiar shapes was superior. Finally, there were no differences among letters, familiar shapes, and non-familiar shapes at the .25-sec. interval. At the .5-sec. interval, performance for familiar shapes was superior to performance for non-familiar shapes. It was concluded that information transfers to short-term visual storage after .25 sec.

Potential applications of neural networks in construction

1992 ◽  
Vol 19 (3) ◽  
pp. 521-529 ◽  
Author(s):  
Osama Moselhi ◽  
Tarek Hegazy ◽  
Paul Fazio
Keyword(s):  
Neural Networks ◽  
Problem Solving ◽  
Expert Systems ◽  
Large Scale ◽  
Computer Applications ◽  
Productivity Level ◽  
The Past ◽  
Intelligent Decision Support Systems ◽  
Potential Applications ◽  
Underlying Mechanisms

During the past decade, several engineering disciplines, including construction, have embarked on developing “intelligent” decision support systems based on artificial intelligence (AI) techniques, including expert systems, symbolic knowledge representation, and logic programming. These systems attempt to capture the domain experts' intelligent behaviour and reasoning process utilized in decision-making, without regard to the underlying mechanisms producing that behaviour. This approach involves describing behaviours, usually with rules and symbols. In contrast, neural networks (NN), another AI-based technique that has been pursued on a large scale during the past few years, does not describe behaviours but rather imitate them. Neural networks are particularly superior to traditional expert systems in providing timely solutions based primarily on analogy with previous experience, rather than reasoning or computation. As such, neural networks have a great potential to work either as a supplement or as a complement to algorithmic and (or) other AI-based systems, providing more suitable tools for solving the industry ill-structured problems.This paper describes several characteristics of neural networks and outlines the advantages and limitations of commonly used NN paradigms. Potential applications of each paradigm in construction are identified. Two example applications are provided to demonstrate the problem-solving capabilities of neural networks: (i) estimation of hourly production rate of an excavation equipment; and (ii) estimation of productivity level for a construction trade. Future possibilities of integrating neural networks with other problem-solving techniques are described. Key words: construction, management techniques, neural networks, expert systems, pattern recognition, computer applications.

Sign in / Sign up

Export Citation Format

Share Document