Artificial neural networks for coordination and control: The portability of experiential representations

It is time to locate connectionist representation theory in the new wave of robotics research. The utility of representations developed in artificial neural networks (ANNs) during learning has been demonstrated in cognitive science research since the 1980s. The research reported here puts learned representations to work in a decentered control task, the disembodied arm problem, in which a mobile robot operates an arm fixed to a table to pick up objects. There is no physical linkage between the arm and the robot and so the robot's point of view must be decentered. This is done by developing a modular Artificial Neural Net system in three stages: (i) a classifier net is trained with laser scan data to output transformationally invariant position classes; (ii) an arm net is trained for picking up objects; (iii) an inter net is trained to communicate and coordinate the sensing and acting. The completed system is shown to create new nonsymbolic transformationally invariant representations in order to perform the effective generalization of decentered viewpoints.     

提高前馈神经网络推广能力的若干实际方法

提高前馈神经网络的推广能力是深受关注的问题，本文根据我们最近提出了一个网络有效推广的准则，从提高网络特征提取能力，分类能力和修改神经元激活函数等几方面给出了若干实际方案，我们在任意手写数字识别问题上的实验结果证实了这些方法的有效性。     

高速网络协议：问题与答案

作者在本文中指出了吉比特网络的目标及其所带来的问题，采用了提出问题并回答问题（或说明可能答案）的方法来阐明。     

An algorithm for architecture partitioning in high-speed networks with bursty traffic

This paper analyzes the performance of a switching architecture. The performance measures include the elapsed time of packet transfer and the waiting time to begin transfer. The architecture is partitioned depending on the type of network used and the expected traffic in the network. Every partition has a switch with a buffer that can absorb surges of bursty traffic within the network partition. The buffer size depends on the type of the network and incoming traffic. The partition size depends on the network bandwidth, network traffic, packet size and buffer size. Examples of different networks are used to show the applications of the model. The results show that the elapsed time of packet switch transfer depends exponentially on the number of partitions in the network.     

Polymer networks based on α,ω-methacrylate-terminated poly(1,3-dioxolane)

α,ω-Methacrylate-terminated poly(1,3-dioxolane)s (polyDXL) were synthesized by cationic ring-opening polymerization of DXL in the presence of methylene-bis(oxyethylmethacrylate) as transfer agent. If the initiator concentration is small compared with the transfer agent concentration, the molecular weights of the polymers are governed by the ratio of the reacted monomer to the reacted transfer agent. The α,ω-methacrylate-terminated polyDXLs obtained undergo free radical polymerization with formation of polyacetal networks. The properties of the networks as function of the molecular weight of the corresponding prepolymers are reported.     

Efficient simulation of hierarchical stochastic activity network models

Stochastic extensions to Petri nets have gained widespread acceptance as a method for describing the dynamic behavior of discrete-event systems. Both simulation and analytic methods have been proposed to solve such models. This paper describes a set of efficient procedures for simulating models that are represented as stochastic activity networks (SANs, a variant of stochastic Petri nets) and composed SAN-based reward models (SBRMs). Composed SBRMs are a hierarchical representation for SANs, in which individual SAN models can be replicated and joined together with other models, in an iterative fashion. The procedures exploit the hierarchical structure and symmetries introduced by the replicate operation in a composed SBRM to reduce the cost of future event list management. The procedures have been implemented as part of a larger performance-dependability modeling package known asUltraSAN, and have been applied to real, large-scale applications. This work was supported in part by the Digital Equipment Corporation Faculty Program: Incentives for Excellence.     

Mixed Mode VLSI Implementation of a Neural Associative Memory

A mixed mode digital/analog special purpose VLSI hardware implementation of an associative memory with neural architecture is presented. The memory concept is based on a matrix architecture with binary storage elements holding the connection weights. To enhance the processing speed analog circuit techniques are applied to implement the algorithm for the association. To keep the memory density as high as possible two design strategies are considered. First, the number of transistors per storage element is kept to a minimum. In this paper a circuit technique that uses a single 6-transistor cell for weight storage and analog signal processing is proposed. Second, the device precision has been chosen to a moderate level to save area as much as possible. Since device mismatch limits the performance of analog circuits, the impact of device precision on the circuit performance is explicitly discussed. It is shown that the device precision limits the number of rows activated in parallel. Since the input vector as well as the output vector are considered to be sparsely coded it is concluded, that even for large matrices the proposed circuit technique is appropriate and ultra large scale integration with a large number of connection weights is feasible.     

人工神经网络预测热油环道启动压力

高含蜡原油管道停输后的再启动问题非常复杂。它涉及因素多，问题的非线性很强。而有足够多神经元的S型三层神经网络理论上可逼近任意有理函数。文章采用误差反向传播网络预测环道实验装置中的启动压力，得到了较好结果。     

Dynamical considerations of network properties

Elasticity is discussed as an aspect of viscoelasticity, which is described by the tube model. The effects of both crosslinks and entanglements contribute to this model and a discussion of how these effects can be quantified is given. At high enough concentration, entanglements ensure the existence of elastic effects even without crosslinks, and a theory is presented on how this dynamical phase change comes about.     

Neural Computation of the MoM Matrix Elements for Planar Configurations

In this paper a new artificial neural network (ANN) based model for the calculation of the method of moments (MoM) matrix elements is presented. Training sets that characterize the matrix elements are first constructed. These sets are then utilized to effectively train two radial basis function (RBF) neural networks to accurately estimate all the elements of the MoM matrix for any mesh used. The potential of the proposed approach is demonstrated in the case of a narrow microstrip line. The current distribution on the microstrip line produced by the trained RBF networks agrees very well with the exact distribution. In addition, the proposed ANN model is much faster than the conventional MoM procedure.