公开(公告)号：US20040181499A1

公开(公告)日：2004-09-16

申请号：US10806501

申请日：2004-03-22

Applicant: Guided Systems Systems Technologies, Inc.

Inventor： J. Eric Corban

IPC: G06F017/00 ,  G06N005/02 ,  G06N003/067 ,  G06N003/063 ,  G06N003/06 ,  G06F015/18 ,  G05B013/02 ,  G06G007/00 ,  G06E003/00 ,  G06E001/00 ,  G06F007/00 ,  G05D003/00 ,  G05D001/00 ,  B64C001/00

CPC classification number: G05B13/027 ,  G05D1/0825

Abstract: A computer system for controlling a nonlinear physical process. The computer system comprises a linear controller and a neural network. The linear controller receives a command signal for control of the nonlinear physical process and a measured output signal from the output of the nonlinear physical process. The linear controller generates a control signal based on the command signal, a measured output signal, and a fixed linear model for the process. The neural network receives the control signal from the linear controller and the measured output signal from the output of the nonlinear physical process. The neural network uses the measured output signal to modify the connection weights of the neural network. The neural network also generates a modified control signal supplied to the linear controller to iterate a fixed point solution for the modified control signal used to control the nonlinear physical process.

Abstract translation: 一种用于控制非线性物理过程的计算机系统。 计算机系统包括线性控制器和神经网络。 线性控制器接收用于控制非线性物理过程的命令信号和来自非线性物理过程的输出的测量输出信号。 线性控制器基于命令信号，测量输出信号和用于该过程的固定线性模型产生控制信号。 神经网络接收来自线性控制器的控制信号和来自非线性物理过程的输出的测量输出信号。 神经网络使用测量的输出信号来修改神经网络的连接权重。 神经网络还生成提供给线性控制器的修改的控制信号，以迭代用于控制非线性物理过程的修改的控制信号的固定点解。

公开(公告)号：US20040107172A1

公开(公告)日：2004-06-03

申请号：US09963971

申请日：2001-09-25

Inventor： Ruibo Wang

IPC: G06N003/067 ,  G06N003/063 ,  G06N003/06 ,  G06E003/00 ,  G06E001/00 ,  G06F015/18

CPC classification number: G06N3/063

Abstract: The present invention discloses low noise, optically coupled optoelectronic and all-optical artificial neuron devices that can be configured in an array to simulate the function of biological neural networks, and methods for making the artificial neurons. In a first optoelectronic embodiment, the device employs the regenerative pulsation property of astable multivibrators as optical pulse generators. Prior art pulse-coupled artificial neurons are subject to undesirable noise interference because the interconnection of such prior art neurons is based on electrical signals conducted through a grid of wires. The present invention obviates the need for hard-wired interconnection of individual neurons in order to cofigure the neurons into a network. In an optoelectronic embodiment, the neuron receives an optical input signal from an external source. A photosensitive detector, disposed in a circuit to control the state of an astable or bistable multivibrator, converts the intensity of the input light into a train of light pulses having a frequency that is a function of the intensity of the input signal. In an all-optical embodiment of an artificial neuron, an input signal is first integrated and the integrated signal transmitted to an optical pulse generator comprised of a nonlinear material disposed within the cavity of a Fabry-Perot etalon. The output of the etalon is a train of light pulses having a frequency that depends upon the intensity of the integrated input signal. When a weak light signal reaches the neuron's input port, there is no light pulse emitted from the output port. By contrast, a strong signal, or a group of weak signals, triggers a short-lived light pulse. The output pulse frequency is a function of the summed input signal power.

Abstract translation: 本发明公开了一种低噪声，光耦合光电子和全光学人造神经元器件，可以配置在阵列中以模拟生物神经网络的功能，以及制造人造神经元的方法。 在第一光电子实施例中，器件采用不稳定多谐振荡器的再生脉动特性作为光脉冲发生器。 现有技术的脉冲耦合人造神经元受到不期望的噪声干扰，因为这种现有技术神经元的互连是基于通过电网格传导的电信号。 本发明消除了对单个神经元的硬线互连的需要，以便将神经元配置到网络中。 在光电子实施例中，神经元从外部源接收光输入信号。 设置在电路中以控制不稳定或双稳态多谐振荡器的状态的光敏检测器将输入光的强度转换成具有作为输入信号的强度的函数的频率的光脉冲串。 在人造神经元的全光学实施例中，首先将输入信号集成，并将积分信号传输到由布置在法布里 - 珀罗标准具的空腔内的非线性材料组成的光脉冲发生器。 标准具的输出是具有取决于积分输入信号的强度的频率的一串光脉冲。 当弱光信号到达神经元的输入端口时，没有从输出端口发出的光脉冲。 相比之下，强信号或一组弱信号触发短暂的光脉冲。 输出脉冲频率是求和输入信号功率的函数。

公开(公告)号：US20040006545A1

公开(公告)日：2004-01-08

申请号：US10189749

申请日：2002-07-03

Inventor： Stanford R. Ovhsinsky

IPC: G06F015/18 ,  G06G007/00 ,  G06N003/067 ,  G06N003/063 ,  G06N003/06 ,  G06E003/00 ,  G06E001/00

CPC classification number: G11C11/54 ,  G06N3/0635 ,  G11C11/5678 ,  G11C13/0004

Abstract: An analog neural computing medium, neuron and neural networks comprising same are disclosed. The neural computing medium includes a phase change material that has the ability to cumulatively respond to multiple synchronous or asynchronous input signals. The introduction of input signals induces transformations among a plurality of accumulation states of the disclosed neural computing medium. The accumulation states are characterized by a high electrical resistance that is substantially identical for all accumulation states. The high electrical resistance prevents the neural computing medium from transmitting signals. Upon cumulative receipt of energy from one or more input signals that equals or exceeds a threshold value, the neural computing medium fires by transforming to a low resistance state that is capable of transmitting signals. The neural computing medium thus closely mimics the neurosynaptic function of a biological neuron. The disclosed neural computing medium may also be configured to perform a weighting function whereby it weights incoming signals and transmits modified signals. The neural computing medium may thus be configured to provide an accumulation function or weighting function and may readily be reconfigured from one function to the other. The disclosed neurons may also include activation units for further transforming signals transmitted by the accumulation units according to a mathematical operation. The artificial neurons, weighting units, accumulation units and activation units may be connected in a variety of ways to form artificial neural networks. Embodiments of several neural networks are disclosed.

Abstract translation: 公开了一种模拟神经计算介质，包含其的神经元和神经网络。 神经计算介质包括具有对多个同步或异步输入信号进行累积响应的能力的相变材料。 输入信号的引入引起所公开的神经计算介质的多个累积状态之间的变换。 累积状态的特征在于对于所有累积状态基本相同的高电阻。 高电阻防止神经计算介质传输信号。 在从等于或超过阈值的一个或多个输入信号累积接收能量时，通过转换为能够发送信号的低电阻状态来激发神经计算媒体。 神经计算介质因此密切地模仿生物神经元的神经突触功能。 所公开的神经计算介质还可以被配置为执行加权功能，由此加权输入信号并发送修改的信号。 因此，神经计算介质可以被配置为提供累积功能或加权函数，并且可以容易地从一个功能重新配置到另一个功能。 所公开的神经元还可以包括用于根据数学运算进一步变换由累积单元发送的信号的激活单元。 人造神经元，加权单元，累积单元和激活单元可以以各种方式连接以形成人造神经网络。 公开了几个神经网络的实施例。

公开(公告)号：US20040030664A1

公开(公告)日：2004-02-12

申请号：US10287963

申请日：2002-11-05

Inventor： Antonios N. Kotoulas ,  Charles Berezin ,  Michael S. Torok ,  Peter F. Lorber

IPC: G06N003/067 ,  G06N003/06 ,  G05B013/02 ,  G06F015/18 ,  G06G007/00 ,  G06E003/00 ,  G06E001/00 ,  G06N003/063

CPC classification number: F16F15/02 ,  B64C2027/7266 ,  F16F2230/18 ,  G10K11/178 ,  Y02T50/34

Abstract: Two neural networks are used to control adaptively a vibration and noise-producing plant. The first neural network, the emulator, models the complex, nonlinear output of the plant with respect to certain controls and stimuli applied to the plant. The second neural network, the controller, calculates a control signal which affects the vibration and noise producing characteristics of the plant. By using the emulator model to calculate the nonlinear plant gradient, the controller matrix coefficients can be adapted by backpropagation of the plant gradient to produce a control signal which results in the minimum vibration and noise possible, given the current operating characteristics of the plant.

公开(公告)号：US20030220889A1

公开(公告)日：2003-11-27

申请号：US10150941

申请日：2002-05-21

Inventor： Bingxue Shi ,  Chun Lu ,  Lu Chen

IPC: G06N003/067 ,  G06N003/063 ,  G06F015/18 ,  G06G007/00 ,  G06E003/00 ,  G06E001/00 ,  G06N003/06

CPC classification number: G06N3/063 ,  G06N3/0635

Abstract: A neural network includes a neuron, an error determination unit, and a weight update unit. The weight update unit includes an analog accumulator. The analog accumulator requires a minimal number of multipliers.

Abstract translation: 神经网络包括神经元，误差确定单元和权重更新单元。 重量更新单元包括模拟累加器。 模拟累加器需要最小数量的乘法器。

公开(公告)号：US20030163436A1

公开(公告)日：2003-08-28

申请号：US10340847

申请日：2003-01-13

Inventor： Jost Seifert

IPC: G06N003/067 ,  G06N003/063 ,  G06N003/06 ,  G06F015/18 ,  G06G007/00 ,  G06E003/00 ,  G06E001/00

CPC classification number: G06N3/0454

Abstract: A neuronal network for modeling an output function that describes a physical system using functionally linked neurons (2), each of which is assigned a transfer function, allowing it to transfer an output value determined from said neuron to the next neuron that is functionally connected to it in series in the longitudinal direction (6) of the network (1), as an input value. The functional relations necessary for linking the neurons are provided within only one of at least two groups (21, 22, 23) of neurons arranged in a transverse direction (7) and between one input layer (3) and one output layer (5). The groups (21, 22, 23) include at least two intermediate layers (11, 12, 13) arranged sequentially in a longitudinal direction (5), each with at least one neuron.

Abstract translation: 一种神经网络，用于对使用功能关联神经元（2）描述物理系统的模型进行建模，每个神经元分配有传递函数，允许其将从所述神经元确定的输出值传递到功能上连接到的神经元 它作为输入值在网络（1）的纵向（6）上串联。 在横向（7）和一个输入层（3）和一个输出层（5）之间布置的神经元的至少两组（21,22,23）中仅提供连接神经元所必需的功能关系， 。 组（21,22,23）包括沿纵向方向（5）顺序布置的至少两个中间层（11,12,13），每个具有至少一个神经元。

公开(公告)号：US20030045951A1

公开(公告)日：2003-03-06

申请号：US09940365

申请日：2001-08-27

Inventor： Alpha Kamchiu Luk

IPC: G06J001/00 ,  G06N003/06 ,  G06N003/063 ,  G06N003/067 ,  G06F015/18 ,  G05B019/42 ,  G05B013/02

CPC classification number: G06K9/6228 ,  G06N99/005

Abstract: A method of identifying features for a classifier includes identifying a set of elements that share a common characteristic, and then identifying a subset of elements within that set which share a second characteristic. Features are then selected that are more commonly possessed by the elements in the subset than the elements in the set but excluding the subset, and that are more commonly possessed by the elements in the set but excluding the subset, as compared to the elements outside the set. A further method of identifying features for a classifier includes defining a list of features, selecting a first feature from that list, identifying a set of elements that possess that first feature, and then identifying a subset of elements within that set which possess any other feature. A feature is then selected that is more commonly possessed by the elements in the subset than the elements in the set but excluding the subset, and that is more commonly possessed by the elements in the set but excluding the subset, as compared to the elements outside the set. If this feature is not already in the list of features, it is added to it. Another feature that has not already been selected is chosen from the list, and the process is repeated using this feature. This continues until every feature in the list of features has been selected.

Abstract translation: 识别分类器的特征的方法包括识别共享共同特征的元素集合，然后识别共享第二特征的该集合内的元素的子集。 然后选择子集中的元素比集合中的元素更常被拥有但不包括子集的特征，并且与集合中的元素相比，集合中的元素更常被拥有但不包括子集。 组。 识别分类器的特征的另一种方法包括定义特征列表，从该列表中选择第一特征，识别拥有该第一特征的一组元素，然后识别具有任何其它特征的该组内的元素的子集 。 然后选择子集中的元素比集合中的元素更常被拥有但不包括子集的特征，并且这是与集合中的元素更常被拥有但不包括子集的特征，与外部的元素相比 集合 如果此功能尚未在功能列表中，则会将其添加到该功能列表中。 从列表中选择另一个尚未选择的功能，并使用此功能重复该过程。 这一直继续，直到功能列表中的每个功能被选中。

公开(公告)号：US20040251941A1

公开(公告)日：2004-12-16

申请号：US10459538

申请日：2003-06-12

Applicant: Winbond Electronics Corporation

Inventor： Bingxue Shi ,  Lu Chen ,  Chun Lu

IPC: H03K009/08 ,  H03K007/08 ,  H03K005/04 ,  H03K003/017 ,  G06F015/18 ,  G06N003/067 ,  G06N003/063 ,  G06N003/06 ,  G06J001/00

CPC classification number: H03K7/08 ,  H03K5/023 ,  H03K5/2472

Abstract: A current-mode pulse-width-modulation (PWM) circuit converts analog current signals into pulse signals. The PWM circuit includes a first I-V converter and one or more second I-V converters, each of the one or more second I-V converters being coupled to one of the current signals. Each of the first and second I-V converters is also coupled to a current generator which generates a current that linearly changes with time. For each of the first and second I-V converters, when a polarity of the input current thereof changes, an output changes between a high voltage level and a low voltage level. A logic circuit is coupled to the first and each second I-V converter to obtain a pulse signal that has a pulse width linearly proportional to the current level of the respective current signal.

Abstract translation: 电流模式脉宽调制（PWM）电路将模拟电流信号转换为脉冲信号。 PWM电路包括第一I-V转换器和一个或多个第二I-V转换器，一个或多个第二I-V转换器中的每一个耦合到电流信号之一。 第一和第二I-V转换器中的每一个还耦合到电流发生器，其产生随时间线性变化的电流。 对于第一和第二I-V转换器中的每一个，当其输入电流的极性改变时，输出在高电压电平和低电压电平之间变化。 逻辑电路耦合到第一和第二I-V转换器，以获得具有与各个电流信号的电流电平成线性比例的脉冲宽度的脉冲信号。

公开(公告)号：US20040064427A1

公开(公告)日：2004-04-01

申请号：US10261265

申请日：2002-09-30

Inventor： Hans R. Depold ,  David John Sirag JR.

IPC: G06N003/063 ,  G06N003/06 ,  G06F015/18 ,  G06G007/00 ,  G06E003/00 ,  G06E001/00 ,  G06N003/067

CPC classification number: G06N3/04

Abstract: A PBNN for isolating faults in a plurality of components forming a physical system comprising a plurality of input nodes each input node comprising a plurality of inputs comprising a measurement of the physical system, and an input transfer function comprising a hyperplane representation of at least one fault for converting the at least one input into a first layer output, a plurality of hidden layer nodes each receiving at least one first layer output and comprising a hidden transfer function for converting the at least one of at least one first layer output into a hidden layer output comprising a root sum square of a plurality of distances of at least one of the at least one first layer outputs, and a plurality of output nodes each receiving at least one of the at least one hidden layer outputs and comprising an output transfer function for converting the at least one hidden layer outputs into an output.

Abstract translation: 一种PBNN，用于隔离形成包括多个输入节点的物理系统的多个部件中的故障，每个输入节点包括多个输入，所述多个输入包括所述物理系统的测量;以及输入传递函数，包括至少一个故障的超平面表示 用于将所述至少一个输入转换成第一层输出;多个隐藏层节点，每个隐藏层节点接收至少一个第一层输出，并且包括隐藏传递函数，用于将至少一个第一层输出中的至少一个转换为隐藏层 输出包括所述至少一个第一层输出中的至少一个的多个距离的根和平方，以及多个输出节点，每个输出节点接收所述至少一个隐层输出中的至少一个，并且包括输出传递函数， 将所述至少一个隐藏层输出转换为输出。

公开(公告)号：US20030225716A1

公开(公告)日：2003-12-04

申请号：US10158067

申请日：2002-05-31

Inventor： Bingxue Shi ,  Guoxing Li

IPC: G06N003/02 ,  G06F015/18 ,  G06N003/06 ,  G06N003/063 ,  G06N003/067

CPC classification number: G06N3/0635 ,  G06N3/063

Abstract: A neural network includes a programmable template matching network and a winner take all network. The programmable template matching network can be programmed with different templates. The WTA network has an output which can be reconfigured and the scale of the WTA network can expanded.

Abstract translation: 神经网络包括可编程模板匹配网络，胜者取所有网络。 可编程模板匹配网络可以使用不同的模板进行编程。 WTA网络具有可重新配置的输出，WTA网络的规模可扩展。