Interval type-2 fuzzy logic and modular neural networks for face recognition applications

In this paper we present a method for response integration in multi-net neural systems using interval type-2 fuzzy logic and fuzzy integrals, with the purpose of improving the performance in the solution of problems with a great volume of information. The method can be generalized for pattern recognition and prediction problems, but in this work we show the implementation and tests of the method applied to the face recognition problem using modular neural networks. In the application we use two interval type-2 fuzzy inference systems (IT2-FIS); the first IT2-FIS was used for feature extraction in the training data, and the second one to estimate the relevance of the modules in the multi-net system. Fuzzy logic is shown to be a tool that can help improve the results of a neural system by facilitating the representation of human perceptions.     

Focusing and polarization in linear,intuitionistic, and classical logics

A focused proof system provides a normal form to cut-free proofs in which the application of invertible and non-invertible inference rules is structured. Within linear logic, the focused proof system of Andreoli provides an elegant and comprehensive normal form for cut-free proofs. Within intuitionistic and classical logics, there are various different proof systems in the literature that exhibit focusing behavior. These focused proof systems have been applied to both the proof search and the proof normalization approaches to computation. We present a new, focused proof system for intuitionistic logic, called LJF, and show how other intuitionistic proof systems can be mapped into the new system by inserting logical connectives that prematurely stop focusing. We also use LJF to design a focused proof system LKF for classical logic. Our approach to the design and analysis of these systems is based on the completeness of focusing in linear logic and on the notion of polarity that appears in Girard’s LC and LU proof systems.     

Introduction: from legal theories to neural networks and fuzzy reasoning

Computational approaches to the law have frequently been characterized as being formalistic implementations of the syllogistic model of legal cognition: using insufficient or contradictory data, making analogies, learning through examples and experiences, applying vague and imprecise standards. We argue that, on the contrary, studies on neural networks and fuzzy reasoning show how AI & law research can go beyond syllogism, and, in doing that, can provide substantial contributions to the law.     

Structural synthesis of fast two-layer neural networks

Methods of construction of structural models of fast two-layer neural networks are considered. The methods are based on the criteria of minimum computing operations and maximum degrees of freedom. Optimal structural models of two-layer neural networks are constructed. Illustrative examples are given. Translated from Kibernetika i Sistemnyi Analiz, No. 4, pp. 47–56, July–August, 2000.     

High-level reasoning,computational challenges for connectionism,and the Conposit solution

Sophisticated symbol processing in connectionist systems can be supported by two primitive representational techniques calledRelative-Position Encoding (RPE) andPattern-Similarity Association (PSA), and a selection technique calledTemporal-Winner-Take-All (TWTA). TWTA effects winner-take-all selection on the basis of fine signal-timing differences as opposed to activation-level differences. Both RPE and PSA are for the encoding of highly temporary associations between representations. RPE is based on the way activation patterns are positioned relative to each other within a network. Under PSA, two patterns are temporarily associated if they have (suitable) subpatterns that are (suitably) similar. The article shows how particular versions of the primitives are used to good effect in a system called Conposit/SYLL. This is a connectionist implementation of a slightly simplified version of a complex existing psychological theory, namely Johnson-Laird's account of syllogistic reasoning. The computational processes in this theory present a major implementational challenge to connectionism. The challenge lies in the mutability, multiplicity, and diversity of the working memory structures, and the elaborateness of the processing needed for them. Conposit/SYLL's techniques allow it to meet the challenge. The implementation of symbolic processing in Conposit/SYLL is an interesting application of connectionism partly because it significantly affects the design of the symbolic processing level itself. In particular, it encourages the use of associative as opposed to pointer-based data structures, and the use of random as opposed to ordered iteration over sets of data structures. In addition, the article discusses Conposit/SYLL's somewhat unusual variable-binding approach.     

基于联合卷积和递归神经网络的交通标志识别

提出了一种联合卷积和递归神经网络的深层网络结构,在卷积神经网络中引入了递归神经网络能学到的组合特征:原始图片先通过一级由k均值聚类学得滤波器的卷积神经网络,得到的结果再同时通过一级卷积和一级递归神经网络,最后得到的特征向量由Softmax分类器进行分类。实验结果表明:在第二级卷积和递归神经网络权重随机的情况下,该网络的识别率已经能够达到98.28%,跟其他网络结构相比,大大减少了训练时间,而且无需复杂的工程技巧。     

基于模糊神经网络的民航物流预测研究与仿真分析

为合理规划我国机场改扩建方案,针对目前民航业特点,从客运量的角度对民航物流预测进行研究,在综合分析影响客运量因素的基础上,提出了模糊对角回归神经网络滚动预测模型.此模型在前端网络处理层对不确定性因素进行模糊量化处理,对确定性因素进行归一化处理,有效地解决了模型输入量纲不一致的问题.通过实际数据的检验与内回归神经网络、外回归神经网络的预测结果相比较,证明应用此模型进行民航客运量预测有较高的预测精度.并在此基础上利用Visual Basic语言开发了民航物流预测仿真系统,对预测结果进行仿真验证,试验结果表明该仿真系统具有广阔的应用前景和推广价值.     

Nominal Reasoning Techniques in Coq: (Extended Abstract)

We explore an axiomatized nominal approach to variable binding in Coq, using an untyped lambda-calculus as our test case. In our nominal approach, alpha-equality of lambda terms coincides with Coq's built-in equality. Our axiomatization includes a nominal induction principle and functions for calculating free variables and substitution. These axioms are collected in a module signature and proved sound using locally nameless terms as the underlying representation. Our experience so far suggests that it is feasible to work from such axiomatized theories in Coq and that the nominal style of variable binding corresponds closely with paper proofs. We are currently working on proving the soundness of a primitive recursion combinator and developing a method of generating these axioms and their proof of soundness from a grammar describing the syntax of terms and binding.     

Neurovision-based logic control of an experimental manufacturing plant using neural net le-net5 and automation Petri nets

In this paper, Petri nets and neural networks are used together in the development of an intelligent logic controller for an experimental manufacturing plant to provide the flexibility and intelligence required from this type of dynamic systems. In the experimental setup, among deformed and good parts to be processed, there are four different part types to be recognised and selected. To distinguish the correct part types, a convolutional neural net le-net5 based on-line image recognition system is established. Then, the necessary information to be used within the logic control system is produced by this on-line image recognition system. Using the information about the correct part types and Automation Petri nets, a logic control system is designed. To convert the resulting Automation Petri net model of the controller into the related ladder logic diagram (LLD), the token passing logic (TPL) method is used. Finally, the implementation of the control logic as an LDD for the real time control of the manufacturing system is accomplished by using a commercial programmable logic controller (PLC).     

Application of artificial neural network and fuzzy logic in a case-based system for initial process parameter setting of injection molding

Determination of initial process meters for injection molding is a highly skilled job and based on skilled operators know-how and intuitive sense acquired through long-term experience rather than a theoretical and analytical approach. Facing with the global competition, the current trial-and-error practice becomes inadequate. In this paper, application of artificial neural network and fuzzy logic in a case-based system for initial process meter setting of injection molding is described. Artificial neural network was introduced in the case adaptation while fuzzy logic was employed in the case indexing and similarity analysis. A computer-aided system for the determination of initial process meter setting for injection molding based on the proposed techniques was developed and validated in a simulation environment. The preliminary validation tests of the system have indicated that the system can determine a set of initial process meters for injection molding quickly without relying on experienced molding personnel, from which good quality molded parts can be produced.     

VLSI implementation of pulsating neural networks

We describe a generic approach for realizing networks of pulsating neurons based on charge pumping of interface states situated in the channel of MOS transistors. Two basic building blocks will be described: the pulse activated charge pumping (PSCP) synapse, and the charge sensitive oscillator (CSO). The PSCP synapse which operates as either a short or a long term memory device which produces a charge packet proportional to the number of pulses applied to its input, will be described in detail together with experimental results demonstrating its capability. The CSO circuit which is a charge controlled oscillator will be described together with simulations of its output frequency dependence on its input voltage, and the relation between the temporal dependence of output waveform on its input charge.     

Classical and superposed learning for quantum weightless neural networks

A supervised learning algorithm for quantum neural networks (QNN) based on a novel quantum neuron node implemented as a very simple quantum circuit is proposed and investigated. In contrast to the QNN published in the literature, the proposed model can perform both quantum learning and simulate the classical models. This is partly due to the neural model used elsewhere which has weights and non-linear activations functions. Here a quantum weightless neural network model is proposed as a quantisation of the classical weightless neural networks (WNN). The theoretical and practical results on WNN can be inherited by these quantum weightless neural networks (qWNN). In the quantum learning algorithm proposed here patterns of the training set are presented concurrently in superposition. This superposition-based learning algorithm (SLA) has computational cost polynomial on the number of patterns in the training set.     

Uniform semantic treatment of default and autoepistemic logics

We revisit the issue of epistemological and semantic foundations for autoepistemic and default logics, two leading formalisms in nonmonotonic reasoning. We develop a general semantic approach to autoepistemic and default logics that is based on the notion of a belief pair and that exploits the lattice structure of the collection of all belief pairs. For each logic, we introduce a monotone operator on the lattice of belief pairs. We then show that a whole family of semantics can be defined in a systematic and principled way in terms of fixpoints of this operator (or as fixpoints of certain closely related operators). Our approach elucidates fundamental constructive principles in which agents form their belief sets, and leads to approximation semantics for autoepistemic and default logics. It also allows us to establish a precise one-to-one correspondence between the family of semantics for default logic and the family of semantics for autoepistemic logic. The correspondence exploits the modal interpretation of a default proposed by Konolige. Our results establish conclusively that default logic can be viewed as a fragment of autoepistemic logic, a result that has been long anticipated. At the same time, they explain the source of the difficulty to formally relate the semantics of default extensions by Reiter and autoepistemic expansions by Moore. These two semantics occupy different locations in the corresponding families of semantics for default and autoepistemic logics.     

Learning in artificial neural systems

This paper presents an overview and analysis of teaming in artificial neural systems (ANSs). It begins with a general introduction to neural networks and connectionist approaches to information processing. The basis for learning in ANSs is then described and compared with classical machine learning. While similar in some ways, ANS learning deviates from tradition in its dependence on the modification of individual weights to bring about changes in a knowledge representation distributed across connections in a network. This unique form of learning is analyzed from two aspects: the selection of an appropriate network architecture for representing the problem, and the choice of a suitable learning rule capable of reproducing the desired function within the given network. The various network architectures are classified, and then identified with explicit restrictions on the types of functions they are capable of representing. The learning rules, i.e., algorithms that specify how the network weights are modified, are similarly taxonomized and, where possible, the limitations inherent to specific classes of rules are outlined.     

A variational approach to intensity approximation for remote sensing images using dynamic neural networks

Abstract: In remote sensing image processing, image approximation, or obtaining a high‐resolution image from a corresponding low‐resolution image, is an ill‐posed inverse problem. In this paper, the regularization method is used to convert the image approximation problem into a solvable variational problem. In regularization, the constraints on smoothness and discontinuity are considered, and the original ill‐posed problem is thereby converted to a well‐posed optimization problem. In order to solve the variational problem, a Hopfield‐type dynamic neural network is developed. This neural network possesses two states that describe the discrepancy between a pixel and adjacent pixels, the intensity evolution of a pixel and two kinds of corresponding weights. Based on the experiment in this study with a Landsat TM image free of added noise and a noisy image, the proposed approach provides better results than other methods. The comparison shows the feasibility of the proposed approach.     

Local cluster neural net: Architecture, training and applications

This paper describes the structure, training and computational abilities of the local cluster (LC) artificial neural net architecture. LC nets are a special class of multilayer perceptrons that use sigmoid functions to generate localised functions. LC nets train as fast as radial basis functions nets and are more general. They are well suited for both, multi-dimensional function approximation and discrete classification. The LC net is the result of our search for a widely applicable neural net architecture suitable for low-cost hardware realisation. The LC net seem particularly well suited for analog VLSI realisation of small-size, low-power, fully parallel neural net chip for real time control applications.     

Non-linear variable selection for artificial neural networks using partial mutual information

Artificial neural networks (ANNs) have been widely used to model environmental processes. The ability of ANN models to accurately represent the complex, non-linear behaviour of relatively poorly understood processes makes them highly suited to this task. However, the selection of an appropriate set of input variables during ANN development is important for obtaining high-quality models. This can be a difficult task when considering that many input variable selection (IVS) techniques fail to perform adequately due to an underlying assumption of linearity, or due to redundancy within the available data.This paper focuses on a recently proposed IVS algorithm, based on estimation of partial mutual information (PMI), which can overcome both of these issues and is considered highly suited to the development of ANN models. In particular, this paper addresses the computational efficiency and accuracy of the algorithm via the formulation and evaluation of alternative techniques for determining the significance of PMI values estimated during selection. Furthermore, this paper presents a rigorous assessment of the PMI-based algorithm and clearly demonstrates the superior performance of this non-linear IVS technique in comparison to linear correlation-based techniques.