The episodal associative memory: Managing manufacturing information on the basis of similarity and associativity

In this paper we report on continuing research on the organization and functionalities of a certain type of computer-implemented associative memory. The associative memory in question is being created to serve as part of a feature-based design system, at present to be used primarily in support of the design, fabrication planning, or inspection planning of discrete mechanical machine parts. This present effort is consonant with prior related work in the realm of case-based reasoning, especially as related to the role of memory in design. Our associative memory innovations are in the use of fuzzy sets and neural net computing in the representation, storage and retrieval of design, fabrication, inspection and materials knowledge. We have designed and implemented a considerable portion of the associative memory and have demonstrated retrieval of previous designs on the basis of qualitative geometry. We have also demonstrated ability to explore materials composition with the objective of meeting critical materials properties constraints.     

A property of neural networks of associative memory with replacing units

A memory capacity exists for artificial neural networks of associative memory. The addition of new memories beyond the capacity overloads the network system and makes all learned memories irretrievable (catastrophic forgetting) unless there is a provision for forgetting old memories. This article describes a property of associative memory networks in which a number of units are replaced when networks learn. In our network, every time the network learns a new item or pattern, a number of units are erased and the same number of units are added. It is shown that the memory capacity of the network depends on the number of replaced units, and that there exists a optimal number of replaced units in which the memory capacity is maximized. The optimal number of replaced units is small, and seems to be independent of the network size. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007     

Phase transitions in associative memory networks

Ideas from random graph theory are used to give an heuristic argument that associative memory structure depends discontinuously on pattern recognition ability. This argument suggests that there may be a certain minimal size for intelligent systems.     

Double-pattern associative memory neural network with pattern loop

A double-pattern associative memory neural network with “pattern loop” is proposed. It can store 2N bit bipolar binary patterns up to the order of 2^2N , retrieve part or all of the stored patterns which all have the minimum Hamming distance with input pattern, completely eliminate spurious patterns, and has higher storing efficiency and reliability than conventional associative memory. The length of a pattern stored in this associative memory can be easily extended from 2N to κN.     

动态核的形态联想记忆网络在灰度图中的应用

介绍了应用于灰度图像的联想记忆和识别的动态核方法,给出了动态核选择的原则和途径。利用动态核可以解决灰度图像在含有随机噪声时的自联想记忆和识别问题,从而给出了一种较好地处理含噪灰度图像恢复的途径。通过实验,验证了该方法的良好性能,取得了较理想的结果。     

Analysis and design of associative memories based on recurrent neural network with discontinuous activation functions

This paper considers a recurrent neural network (RNN) with a special class of discontinuous activation function which is piecewise constants in the state space. One sufficient condition is established to ensure that the novel recurrent neural networks can have (4k−1)ⁿ locally exponential stable equilibrium points. Such RNN is suitable for synthesizing high-capacity associative memories. The design procedure is presented with the method of singular value decomposition. Finally, the validity and performance of the results are illustrated by use of two numerical examples.     

Pattern classification using a linear associative memory

Pattern classification is a very important image processing task. A typical pattern classification algorithm can be broken into two parts; first, the pattern features are extracted and, second, these features are compared with a stored set of reference features until a match is found. In the second part, usually one of the several clustering algorithms or similarity measures is applied. In this paper, a new application of linear associative memory (LAM) to pattern classification problems is introduced. Here, the clustering algorithms or similarity measures are replaced by a LAM matrix multiplication. With a LAM, the reference features need not be separately stored. Since the second part of most classification algorithms is similar, a LAM standardizes the many clustering algorithms and also allows for a standard digital hardware implementation. Computer simulations on regular textures using a feature extraction algorithm achieved a high percentage of successful classification. In addition, this classification is independent of topological transformations.     

Chaos control and associative memory of a time-delay globally coupled neural network using symmetric map

A chaotic neural network called time-delay globally coupled neural network using symmetric map (TDSG) is proposed for information processing applications. Firstly, its rich dynamic behaviors are exhibited and the output stability is demonstrated by using a parameter modulated control method. Secondly, the associative memory of TDSG is investigated by the control method. It is observed that the stable output sequence only contains stored pattern and its reverse pattern and the TDSG finally converges to the stored pattern which has the smallest Hamming distance to the initial patterns with noise. At last, strong information recovery ability of the TDSG is illustrated by comparative experiments.     

Financial distress prediction using the hybrid associative memory with translation

This paper presents an alternative technique for financial distress prediction systems. The method is based on a type of neural network, which is called hybrid associative memory with translation. While many different neural network architectures have successfully been used to predict credit risk and corporate failure, the power of associative memories for financial decision-making has not been explored in any depth as yet. The performance of the hybrid associative memory with translation is compared to four traditional neural networks, a support vector machine and a logistic regression model in terms of their prediction capabilities. The experimental results over nine real-life data sets show that the associative memory here proposed constitutes an appropriate solution for bankruptcy and credit risk prediction, performing significantly better than the rest of models under class imbalance and data overlapping conditions in terms of the true positive rate and the geometric mean of true positive and true negative rates.     

Non-uniform cellular automata based associative memory: Evolutionary design and basins of attraction

This paper presents the synthesis and analysis of a special class of non-uniform cellular automata (CAs) based associative memory, termed as generalized multiple attractor CAs (GMACAs). A reverse engineering technique is presented for synthesis of the GMACAs. The desired CAs are evolved through an efficient formulation of genetic algorithm coupled with the reverse engineering technique. This has resulted in significant reduction of the search space of the desired GMACAs. Characterization of the basins of attraction of the proposed model establishes the sparse network of GMACAs as a powerful pattern recognizer for memorizing unbiased patterns. Theoretical analysis also provides an estimate of the noise accommodating capability of the proposed GMACA based associative memory. An in-depth analysis of the GMACA rule space establishes the fact that more heterogeneous CA rules are capable of executing complex computation like pattern recognition.     

A dynamic associative memory system by adopting an amygdala model

Although several kinds of computational associative memory models and emotion models have been proposed since the last century, the interaction between memory and emotion is almost always neglected in these conventional models. This study constructs a dynamic memory system, named the amygdala-hippocampus model, which intends to realize dynamic auto-association and the mutual association of time-series patterns more naturally by adopting an emotional factor, i.e., the functional model of the amygdala given by Morén and Balkenius. The output of the amygdala is designed to control the recollection state of multiple chaotic neural networks (MCNN) in CA3 of the hippocampus-neocortex model proposed in our early work. The efficiency of the proposed association system is verified by computer simulation using several benchmark time-series patterns. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

An adaptive associative memory system based on autonomous reaction between image memories

To construct a “thinking-like” processing system, a new architecture of an adaptive associative memory system is proposed. This memory system treats “images” as basic units of information, and adapts to the environment of the external world by means of autonomous reactions between the images. The images do not have to be clear, distinct symbols or patterns; they can be ambiguous, indistinct symbols or patterns as well. This memory system is a kind of neural network made up of nodes and links called a localist spreading activation network. Each node holds one image in a localist manner. Images in high-activity nodes interact autonomously and generate new images and links. By this reaction between images, various forms of images are generated automatically under constraints of links with adjacent nodes. In this system, three simple image reaction operations are defined. Each operation generates a new image by combining pseudofigures or features and links of two images. This work was presented, in part, at the Fourth International Symposium on Artificial Life and Robotics, Oita, Japan, January 19–22, 1999     

Operation of an associative memory

This paper describes the operation of an associative memory (LYAM) governed by only ordinary differential equations, useful for pattern clustering. Several computer simulations illustrate its operation as an unsupervised classifier, vector quantizer, and content-addressable memory.     

Comparison of the Prim-Dijkstra and kraskal algorithms on an associative parallel processor

With the help of a model of an associative parallel processor with vertical processing (STAR computer), Prim-Dijkstra and Kraskal algorithms for finding a minimal spanning tree of an undirected graph represented in the form of a list of edges and their weights are compared. A relatively simple representation of the Prim-Dijkstra algorithm is constructed in which the initial node is taken into account. The Kraskal algorithm is also presented and the possibility of eliminating the stage of preliminary sorting of edges by their weights is shown. Translated from Kibernetika i Sistemnyi Analiz, No. 2. pp. 19–27, March–April, 2000.     

Design of a multitask neurovision processor

The architecture of a biologically motivated visual-information processor that can perform a variety of tasks associated with the early stages of machine vision is described. The computational operations performed by the processor emulate the spatiotemporal information-processing capabilities of certain neural-activity fields found along the human visual pathway. The state-space model of the neurovision processor is a two-dimensional nural network of densely interconnected nonlinear processing elements PE's. An individual PE represents the dynamic activity exhibited by a spatially localized population of excitatory and inhibitory nerve cells. Each PE may receive inputs from an external signal space as well as from the neighboring PE's within the network. The information embedded within the signal space is extracted by the feedforward subnet. The feedback subnet of the neurovision processor generates useful steady-state and temporal-response characteristics that can be used for spatiotemporal filtering, short-term visual memory, spatiotemporal stabilization, competitive feedback interaction, and content-addressable memory. To illustrate the versatility of the multitask processor design for machine-vision applications, a computer simulation of a simplified vision system for filtering, storing, and classifying noisy gray-level images in presented.     

Color image associative memory on a class of Cohen-Grossberg networks

In this paper, neural associative memories for storing gray-scale and true color images are presented based on a class of reduced Cohen-Grossberg neural networks. Some fundamental conditions for endowing the networks with retrieval properties are proposed. Moreover, a system designing procedure is developed by using matrix decomposition. Numerical simulations show that the constructed networks can act as reliable noise-reducing systems for storing and retrieving color images.     

基于 MATLAB GUI 的数字图像处理仿真系统设计*

数字图像处理理论性较强,学习具体算法时需要形象化演示。MATLAB提供了图像处理工具箱,能够快速构建算法演示;GUI更提供了友好的图形用户交互界面。论文依托 MATLAB GUI ,实现数字图像处理算法的仿真系统。该系统具有良好的交互性和可扩展性。实践效果表明,提高了学生学习数字图像处理的兴趣和效果。     

Comparisons of air traffic control implementations on an associative processor with a MIMD and consequences for parallel computing

This paper has two complementary focuses. The first is the system design and algorithmic development for air traffic control (ATC) using an associative SIMD processor (AP). The second is the comparison of this implementation with a multiprocessor implementation and the implications of these comparisons. This paper demonstrates how one application, ATC, can more easily, more simply, and more efficiently be implemented on an AP than is generally possible on other types of traditional hardware. The AP implementation of ATC will take advantage of its deterministic hardware to use static scheduling. The software will be dramatically smaller and cheaper to create and maintain. Likewise, a large AP system will be considerably simpler and cheaper than the MIMD hardware currently used. While APs were used for ATC-type applications earlier, these are no longer available. We use a ClearSpeed CSX600 accelerator to emulate the AP solutions of ATC on an ATC prototype consisting of eight data-intensive ATC real-time tasks. Its performance is compared with an 8-core multiprocessor (MP) using OpenMP. Our extensive experiments show that the AP implementation meets all deadlines while the MP will regularly miss a large number of deadlines. The AP code will be similar in size to sequential code for the same tasks and will avoid all of the additional support software needed with an MP to handle dynamic scheduling, load balancing, shared resource management, race conditions, false sharing, etc. At this point, essentially only MIMD systems are built. Many of the advantages of using an AP to solve an ATC problem would carry over to other applications. AP solutions for a wide variety of applications will be cited in this paper. Applications that involve a high degree of data parallelism such as database management, text processing, image processing, graph processing, bioinformatics, weather modeling, managing UAS (Unmanned Aircraft Systems or drones) etc., are good candidates for AP solutions. This raises the issue of whether we should routinely consider using non-multiprocessor hardware like the AP for applications where substantially simpler software solutions will normally exist. It also raises the question of whether the use of both AP and MIMD hardware in a single hetergeneous system could provide more versatility and efficiency. Either the AP or MIMD could serve as the primary system, but could hand off jobs it could not handle efficiently to the other system.     

Processor-time optimal parallel algorithms for digitized images on mesh-connected processor arrays

We present processor-time optimal parallel algorithms for several problems onn ×n digitized image arrays, on a mesh-connected array havingp processors and a memory of sizeO(n ²) words. The number of processorsp can vary over the range [1,n ^3/2] while providing optimal speedup for these problems. The class of image problems considered here includes labeling the connected components of an image; computing the convex hull, the diameter, and a smallest enclosing box of each component; and computing all closest neighbors. Such problems arise in medium-level vision and require global operations on image pixels. To achieve optimal performance, several efficient data-movement and reduction techniques are developed for the proposed organization.This research was supported in part by the National Science Foundation under Grant IRI-8710836 and in part by DARPA under Contract F33615-87-C-1436 monitored by the Wright Patterson Airforce Base.     

地形仿真中遥感影像的应用及处理技术

在分析几种地球表面表示方法的优缺点的基础上,提出用遥感影像作为三维地形仿真中表面纹理图以提高真实感的思想;提出不同高度实体视景地形的仿真应采用多级精度仿真的方案,并分析了地形仿真精度与影像分辨率间的关系;详细叙述了该应用方案中影像数据处理的流程及方法,包括：数据准备、真彩波段组合、影像增强、几何校正、拼接镶嵌、投影变换及三维地形生成等。实践证明,合理的影像数据处理方法和步骤可减少工作量并保证影像处理的精度和效果,现势件好,真彩组合的遥感影像用作三维地形仿真的表面纹理图是提高仿真真实感的有效技术途径。