A defence of functional modularity

Although belief in the existence of mental modules of some form is widespread among cognitive researchers, neurally sophisticated researchers commonly resist the view that cognitive processing involves modules that are functionally independent of one another. Moreover, within the past few years, at least three noted researchers (Fodor, Kosslyn and Uttal) have called into serious question the existence of distinct modules in broad areas of human cognition. This paper offers a defence of the existence of functionally independent modules, which, though spatially distributed, communicate via traditionally conceived input/output channels. This defence proceeds: (i) by showing that the anti-modularity arguments of Fodor, Kosslyn and Uttal are not compelling; (ii) by presenting theoretically-grounded reasons why any connectionist is committed, via the most basic tenets of connectionism, to accepting the existence of functionally independent modules; (iii) by presenting holistically inclined connectionists with a novel challenge, namely, to demonstrate that a single, holistic network could display strong levels of generalization as a side-effect of multiple, previously acquired skills. In the course of these arguments, I examine a recent generalization challenge posed by Phillips (2000, Connection Science, 12: 1–19) to eliminative connectionists.     

Detection and classification of flaws in concrete structure using bispectra and neural networks

The problem addressed in this paper is the detection and classification of flaws in concrete structure. It is known that higher-order spectra contain information not present in the power spectrum and can suppress Gaussian noise. Thus estimates of higher-order spectra have been shown to be useful in certain signal processing problems. This paper is concerned with the feature extraction from bispectra for concrete flaw detection. Impact-echo experiments are carried out for three different types of flaw in concrete structure. For each monitoring signal, after bispectral estimation, features are selected from the modules of bispectra in the primary region. For automatic interpretation, a multilayer back-propagation neural network is used as a classifier. Both clean data and data with additive white Gaussian noise are used for training and testing. The classification results obtained experimentally demonstrate that this method has good detection rates in varying environments.     

Evolving unipolar memristor spiking neural networks

Neuromorphic computing – brain-like computing in hardware – typically requires myriad complimentary metal oxide semiconductor spiking neurons interconnected by a dense mesh of nanoscale plastic synapses. Memristors are frequently cited as strong synapse candidates due to their statefulness and potential for low-power implementations. To date, plentiful research has focused on the bipolar memristor synapse, which is capable of incremental weight alterations and can provide adaptive self-organisation under a Hebbian learning scheme. In this paper, we consider the unipolar memristor synapse – a device capable of non-Hebbian switching between only two states (conductive and resistive) through application of a suitable input voltage – and discuss its suitability for neuromorphic systems. A self-adaptive evolutionary process is used to autonomously find highly fit network configurations. Experimentation on two robotics tasks shows that unipolar memristor networks evolve task-solving controllers faster than both bipolar memristor networks and networks containing constant non-plastic connections whilst performing at least comparably.     

The cultural evolution of communication in a population of neural networks

Human language is learned, symbolic and exhibits syntactic structure, a set of properties which make it unique among naturally-occurring communication systems. How did human language come to be as it is? Language is culturally transmitted and cultural processes may have played a role in shaping language. However, it has been suggested that the cultural transmission oflanguage is constrained by some language-specific innate endowment. The primary objective of the research outlined in this paper is to investigate how such an endowment would influence the acquisition of langage and the dynamics of the repeated cultural transmission of language. To this end, a new connectionist model of the cultural evolution of communication is presented. In this model an individual's innate endowment is considered to be a learning rule with an associated learning bias. The model allows manipulations to be made to this learning apparatus andthe impact of such manipulations on the processes of language acquisition and language evolution to be explored. These investigations reveal that an innate endowment consisting of an ability to read the communicative intentions of others and a bias towards acquiring one-to-one mappings between meanings and signals results in the emergence, through purely cultural processes, of optimal communication. It has previously been suggested that humans possess just such an innate endowment. Properties of human language may therefore best be explained in terms of cultural evolution on an innate substrate.     

Reuse of information in multi-layer sigma-pi neural networks

This new work is an extension of existing research into artificial neural networks (Neville and Stonham, Connection Sci.: J. Neural Comput. Artif. Intell. Cognitive Res., 7, pp. 29–60, 1995; Neville, Neural Net., 45, pp. 375–393, 2002b). These previous studies of the reuse of information (Neville, IEEE World Congress on Computational Intelligence, 1998b, pp. 1377–1382; Neville and Eldridge, Neural Net., pp. 375–393, 2002; Neville, IEEE World Congress on Computational Intelligence, 1998c, pp. 1095–1100; Neville, IEEE 2003 International Joint Conference on Neural Networks, 2003; Neville, IEEE IJCNN'04, 2004 International Joint Conference on Neural Networks, 2004) are associated with a methodology that prescribes the weights, as opposed to training them. In addition, they work with smaller networks. Here, this work is extended to include larger nets. This methodology is considered in the context of artificial neural networks: geometric reuse of information is described mathematically and then validated experimentally. The theory shows that the trained weights of a neural network can be used to prescribe the weights of other nets of the same architecture. Hence, the other nets have prescribed weights that enable them to map related geometric functions. This means the nets are a method of ‘reuse of information’. This work is significant in that it validates the statement that, ‘knowledge encapsulated in a trained multi-layer sigma-pi neural network (MLSNN) can be reused to prescribe the weights of other MLSNNs which perform similar tasks or functions’. The important point to note here is that the other MLSNNs weights are prescribed in order to represent related functions. This implies that the knowledge encapsulated in the initially trained MLSNN is of more use than may initially appear.     

Self-refreshing memory in artificial neural networks: learning temporal sequences without catastrophic forgetting

While humans forget gradually, highly distributed connectionist networks forget catastrophically: newly learned information often completely erases previously learned information. This is not just implausible cognitively, but disastrous practically. However, it is not easy in connectionist cognitive modelling to keep away from highly distributed neural networks, if only because of their ability to generalize. A realistic and effective system that solves the problem of catastrophic interference in sequential learning of ‘static’ (i.e. non-temporally ordered) patterns has been proposed recently (Robins 1995 Robins, AV. 1995. Catastrophic forgetting, rehearsal and pseudorehearsal. Connection Science, 7: 123–146. [Taylor & Francis Online] [Google Scholar], Connection Science, 7: 123–146, 1996, Connection Science, 8: 259–275, Ans and Rousset 1997 Ans, B and Rousset, S. 1997. Avoiding catastrophic forgetting by coupling two reverberating neural networks. CR Académie des Sciences Paris, Life Sciences, 320: 989–997.  [Google Scholar], CR Académie des Sciences Paris, Life Sciences, 320: 989–997, French 1997 French, RM. 1997. Pseudo-recurrent connectionist networks: an approach to the ‘sensitivity–stability’ dilemma. Connection Science, 9: 353–379. [Taylor & Francis Online] [Google Scholar], Connection Science, 9: 353–379, 1999, Trends in Cognitive Sciences, 3: 128–135, Ans and Rousset 2000 Ans, B and Rousset, S. 2000. Neural networks with a self-refreshing memory: knowledge transfer in sequential learning tasks without catastrophic forgetting. Connection Science, 12: 1–19. [Taylor & Francis Online], [Web of Science ®] [Google Scholar], Connection Science, 12: 1–19). The basic principle is to learn new external patterns interleaved with internally generated ‘pseudopatterns’ (generated from random activation) that reflect the previously learned information. However, to be credible, this self-refreshing mechanism for static learning has to encompass our human ability to learn serially many temporal sequences of patterns without catastrophic forgetting. Temporal sequence learning is arguably more important than static pattern learning in the real world. In this paper, we develop a dual-network architecture in which self-generated pseudopatterns reflect (non-temporally) all the sequences of temporally ordered items previously learned. Using these pseudopatterns, several self-refreshing mechanisms that eliminate catastrophic forgetting in sequence learning are described and their efficiency is demonstrated through simulations. Finally, an experiment is presented that evidences a close similarity between human and simulated behaviour.     

初始组织特征对充型过程中初生相演变的影响

采用电磁搅拌法制备具有不同微观初始组织特征的半固态合金熔体,利用半固态挤压铸造法铸造螺旋线试样,使用定量金相技术分析试样的初始组织、成形的螺旋线试样不同长度上的初生相微观组织特征参数(固相率、晶粒尺寸和形状因子),研究半固态合金熔体充型过程中初生相组织的演变规律.结果表明:初生固相率在充型的沿程流动过程中变化较小;初生相的晶粒尺寸、形状因子沿充型长度呈现波浪形变化,波峰和波谷出现的位置与充型长度没有明确的关系;充型后初生相晶粒尺寸的变化幅度与充型前半固态合金熔体初生相晶粒尺寸的大小有对应关系.     

Tracking control of two-wheel driven mobile robot using compound sine function neural networks

The purpose of this paper is to propose a compound sine function neural network (NN) with continuous learning algorithm for the velocity and orientation angle tracking control of a mobile robot. Herein, two NN controllers embedded in the closed-loop control system are capable of on-line continuous learning and do not require any knowledge of the dynamics model. The neuron function of the hidden layer in the three-layer feed-forward network structure is on the basis of combining a sine function with a unipolar sigmoid function. In the NN algorithm, the weight values are only adjusted between the nodes in hidden layer and the output nodes, while the weight values between the input layer and the hidden layer are one, that is, constant, without the weight adjustment. The developed NN controllers have simple algorithm and fast learning convergence. Therefore, the proposed NN controllers can be suitable for the real-time tracking control of the mobile robots. The simulation results show that the proposed NN controller has better control performance in the tracking control of the mobile robot. The compound sine function NN provides a new way to solve tracking control problems for a mobile robot.     

再熔融加热对亚共晶铝硅合金半固态组织演变的影响

提出一种新的亚共晶铝硅合金半固态坯料制备方法，即再熔融加热制备法，其特点是不对熔融金属施加任何机械的或电磁的搅拌力，而是通过在金属凝固过程中进行再熔融加热，获得具有近球状初晶相的亚共晶铝硅合金半固态组织。研究了Al-7％Si—0．2％Ti半固态合金再熔融加热制备过程中再熔融温度及保温时间等关键工艺参数对半固态组织的影响，并讨论了再熔融加热对半固态组织演变的影响。     

An evaluation of recurrent neural network modelling for the prediction of damage evolution during forming

This paper examines the efficiency and capability of Dynet, a recurrent neural network model for the prediction of the damage evolution during hot non-uniform, non-isothermal forging on the basis of a limited number of damage snapshots during the process. A Bayesian algorithm is introduced to optimise the hyperparameters related to the noise level and weight decay. In order to examine the capability of the model to capture the underlying trends when presented with sparse and noisy evidence, a synthetic relation between damage accumulation in a metal matrix composite and strain, strain rate and deformation temperature has been used to generate training data (evidence) of varying accuracy and sparseness. The results show that the Bayesian algorithm performs very well, and that no significant overfitting is observed. In addition, this algorithm not only gives the expectation value of damage level, but also an estimate of its uncertainty.     

钦州那雾山地质遗迹形成演化过程及保护对策研究

文章以钦州那雾山地质遗迹为例,通过实地综合科学考察、系统整理、归纳分析、定性评价等,初步查明了钦州那雾山地质遗迹特征,分析了地质遗迹形成演化过程,提出了地质遗迹保护对策,为后续地质遗迹保护工作提供参考.     

人工神经网络在过程工业中的应用

当前,集过程实时监测、故障诊断、模拟、优化、控制以及调度等各层次功能于一体的过程工业生产过程综合自动化成了过程工业界和学术界共同关注的热点之一.与离散产品的制造业相比,由于流程型工业过程具有强非线性的特点,给实现流程工业综合自动化造成很大的困难,因此必须引入新的思路,开发新的方法.人工神经网络是一种模拟人类思维活动的并行分布式的信息处理系统,可用于映射任何连续函数及进行模式识别,同时还具有自学习功能,实现知识的自动获取,自20世纪90年代以来已在过程系统工程领域内受到广泛的瞩目.重点讨论了人工神经网络在过程系统建模、故障诊断以及在线优化等方面的应用,以展示这种方法在流程工业综合自动化中的良好应用前景.     

用于大面积芯片互连的纳米银膏无压烧结行为

为研究烧结纳米银材料细观孔隙结构对宏观力学性能的影响,首先利用高斯滤波算法和分位数切割函数生成具有不同孔隙率(0.1,0.2和0.3)的代表性微元(RVE). 通过对RVE施加周期性边界条件,获得其单轴拉伸力学性能,使用Abaqus软件建立了由烧结纳米银颗粒制成的搭接接头的宏观模型模拟剪切试验,烧结层的材料属性与预测的RVE的弹塑性应力-应变曲线保持一致. 结果表明,随着孔隙率的减小,RVE模型的弹性模量和屈服强度增大;值得注意的是,随着应变的增大,塑性变形最后阶段的应力呈现较大的下降趋势,使得材料更容易受到损伤. 通过比较宏观模型的剪切模拟,可以观察到孔隙率的变化对烧结纳米银颗粒的剪切变形有显著影响,具体而言,随着孔隙率的增加,孔隙部位更容易出现裂纹并扩展,形成多个孔隙的贯通裂纹,从而导致烧结银的抗剪强度降低.

     

Context-free and context-sensitive dynamics in recurrent neural networks

Continuous-valued recurrent neural networks can learn mechanisms for processing context-free languages. The dynamics of such networks is usually based on damped oscillation around fixed points in state space and requires that the dynamical components are arranged in certain ways. It is shown that qualitatively similar dynamics with similar constraints hold for aⁿbⁿcⁿ , a context-sensitive language. The additional difficulty with aⁿbⁿcⁿ , compared with the context-free language aⁿbⁿ , consists of 'counting up' and 'counting down' letters simultaneously. The network solution is to oscillate in two principal dimensions, one for counting up and one for counting down. This study focuses on the dynamics employed by the sequential cascaded network, in contrast to the simple recurrent network, and the use of backpropagation through time. Found solutions generalize well beyond training data, however, learning is not reliable. The contribution of this study lies in demonstrating how the dynamics in recurrent neural networks that process context-free languages can also be employed in processing some context-sensitive languages (traditionally thought of as requiring additional computation resources). This continuity of mechanism between language classes contributes to our understanding of neural networks in modelling language learning and processing.     

Artificial neural networks in steel-mushy aluminum pressing bonding

1　ＩＮＴＲＯＤＵＣＴＩＯＮＦｏｒｓｔｅｅｌ ａｌｕｍｉｎｕｍｂｏｎｄｉｎｇ ,ｉｆａｌｕｍｉｎｕｍｓｏｌｉｄｆｒａｃｔｉｏｎｉｓ 10 0 % ,ｉｔｉｓｓｔｅｅｌ ａｌｕｍｉｎｕｍｓｏｌｉｄｔｏｓｏｌｉｄｂｏｎｄｉｎｇ ;ｉｆａｌｕｍｉｎｕｍｓｏｌｉｄｆｒａｃｔｉｏｎｉｓ 0 ,ｉｔｉｓｓｔｅｅｌ ａ ｌｕｍｉｎｕｍｓｏｌｉｄｔｏｌｉｑｕｉｄｂｏｎｄｉｎｇ ;ｉｆａｌｕｍｉｎｕｍｓｏｌｉｄｆｒａｃｔｉｏｎｉｓｗｉｔｈｉｎ 0～ 10 0 % ,ｔｈｅｂｏｎｄｉｎｇｉｓｓｔｅｅｌ ｍｕｓｈｙａｌｕｍｉｎｕｍｂｏｎｄｉｎｇ .Ｆｏｒ…     

Stack-like and queue-like dynamics in recurrent neural networks

What dynamics do simple recurrent networks (SRNs) develop to represent stack-like and queue-like memories? SRNs have been widely used as models in cognitive science. However, they are interesting in their own right as non-symbolic computing devices from the viewpoints of analogue computing and dynamical systems theory. In this paper, SRNs are trained on two prototypical formal languages with recursive structures that need stack-like or queue-like memories for processing, respectively. The evolved dynamics are analysed, then interpreted in terms of simple dynamical systems, and the different ease with which SRNs aquire them is related to the properties of these simple dynamical systems. Within the dynamical systems framework, it is concluded that the stack-like language is simpler than the queue-like language, without making use of arguments from symbolic computation theory.     

Emergent latent symbol systems in recurrent neural networks

Fodor and Pylyshyn [(1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28(1–2), 3–71] famously argued that neural networks cannot behave systematically short of implementing a combinatorial symbol system. A recent response from Frank et al. [(2009). Connectionist semantic systematicity. Cognition, 110(3), 358–379] claimed to have trained a neural network to behave systematically without implementing a symbol system and without any in-built predisposition towards combinatorial representations. We believe systems like theirs may in fact implement a symbol system on a deeper and more interesting level: one where the symbols are latent – not visible at the level of network structure. In order to illustrate this possibility, we demonstrate our own recurrent neural network that learns to understand sentence-level language in terms of a scene. We demonstrate our model's learned understanding by testing it on novel sentences and scenes. By paring down our model into an architecturally minimal version, we demonstrate how it supports combinatorial computation over distributed representations by using the associative memory operations of Vector Symbolic Architectures. Knowledge of the model's memory scheme gives us tools to explain its errors and construct superior future models. We show how the model designs and manipulates a latent symbol system in which the combinatorial symbols are patterns of activation distributed across the layers of a neural network, instantiating a hybrid of classical symbolic and connectionist representations that combines advantages of both.     

Effects of compressing and remelting in SIMA processing on semi-solid structure evolution of an Al-Zn wrought alloy

Structure evolution of an Al-Zn wrought alloy in remelting processing in the strain induced melt activated (SIMA) semi-solid procedure was observed, and effects of factors, the remelting temperature, the holding time, and the compression strain, on structures and grain sizes of the alloy were investigated. The results show that (1) the proper temperature of remelting is in the range of 610 to 615 ℃; (2) the grain size in specimen with greater compression strain is smaller than that with smaller compression strain in condition of the same remelting temperature and holding time, and the grain size in local area with great local equivalent strain is smaller than that with small one; (3) liquid occurs in form of cluster in matrix during remelting and its quantity increases with remelting time increasing; liquid in specimen with great compression strain occurs earlier than that with small one, and quantity of liquid in the center of specimen with greater local equivalent strain is greater man that in the tw     

300M钢的热变形行为及其变形组织演变研究

基于热压缩实验,对300M钢在应变速率为10s-1下的热变形行为及其变形组织演变进行了研究。结果表明:在试样高度压下量为50%,变形温度为700～750℃时,300M钢的应力-应变曲线呈流变失稳型,且变形组织出现绝热剪切;当变形温度为800～1000℃时,300M钢的应力-应变曲线呈双峰不连续动态再结晶型,且热变形过程出现了两轮动态再结晶;当变形温度为1050～1180℃时,300M钢的应力-应变曲线呈单峰不连续动态再结晶型,且热变形过程只发生了一轮动态再结晶。