公开(公告)号：US06735579B1

公开(公告)日：2004-05-11

申请号：US09477638

申请日：2000-01-05

申请人： Roger L. Woodall

发明人： Roger L. Woodall

IPC分类号： G06E100

CPC分类号： G06N3/049 ,  G06K9/6232

摘要： A static memory processor for pattern recognition and an input data dimensionality reduction is provided having a multi-layer harmonic neural network and a classifier network. The multi-layer harmonic neural network receives a fused feature vector of the pattern to be recognized from a neural sensor and generates output vectors which aid in discrimination between similar patterns. The fused feature vector and each output vector are separately provided to corresponding positional king of the mountain (PKOM) circuits within the classifier network. Each PKOM circuit generates a positional output vector with only the element corresponding to the element of the fused feature vector or output vector having the highest contribution in its respective vector having a value corresponding to one. The positional output vectors are active in a multidimensional memory space and are read by a recognition vector array which generates class likelihood outputs determined by the occupied memory space. The class likelihood outputs are provided to a class PKOM circuit which outputs classification identifiers to provide the desired pattern recognition.

摘要翻译： 提供了一种用于模式识别和输入数据维数降低的静态存储器处理器，其具有多层谐波神经网络和分类器网络。 多层谐波神经网络从神经传感器接收要识别的图案的融合特征向量，并产生有助于区分相似图案的输出向量。 融合特征向量和每个输出向量分别提供给分类器网络内山（PKOM）电路的相应位置王。 每个PKOM电路产生位置输出向量，其中只有对应于具有对应于1的值的相应向量中具有最高贡献的融合特征向量或输出向量的元素的元素。 位置输出向量在多维存储器空间中是有效的，并且由识别向量阵列读取，所述识别向量阵列产生由占用的存储器空间确定的类似性输出。 类似性输出被提供给类别PKOM电路，其输出分类标识符以提供期望的模式识别。

公开(公告)号：US06625585B1

公开(公告)日：2003-09-23

申请号：US09549746

申请日：2000-04-14

申请人： John D. MacCuish ,  Christodoulos A. Nicolaou

发明人： John D. MacCuish ,  Christodoulos A. Nicolaou

IPC分类号： G06E100

CPC分类号： G06F19/707 ,  C40B30/02

摘要： A system for helping a chemist to identify pharmacophoric mechanisms, based on a set of input data representing many chemical compounds. Given an input data set defining for each compound a feature characteristic and an activity characteristic, a computer agglomeratively clusters representations of the molecules based on their feature characteristics. The result of this process is a multi-domain pyramid structure, made up of a number of nodes each representing one or more molecules. For each node, the computer identifies a representative feature set (such as a largest substructure common among the molecules in the node) and a representative activity level (such as an average of the activity levels of the molecules in the node). The computer then provides as output to a chemist a description of all or part of the pyramid. This process thus converts a large set of raw data into an understandable and commercially useful form, which can assist the chemist in developing beneficial new pharmaceuticals.

摘要翻译： 基于一组代表许多化合物的输入数据，帮助化学家鉴定药效机制的系统。 给定为每个化合物定义特征特征和活动特征的输入数据集，计算机基于它们的特征特征来聚集分子的表示。 该过程的结果是多畴金字塔结构，由多个节点组成，每个节点表示一个或多个分子。 对于每个节点，计算机识别代表性特征集（例如在节点中的分子中共同的最大子结构）和代表性活动水平（例如节点中分子的活动水平的平均值）。 然后，计算机向化学家提供全部或部分金字塔的描述。 因此，该过程将大量原始数据转化为可理解和商业有用的形式，这可以帮助化学家开发有益的新药。

公开(公告)号：US06269323B1

公开(公告)日：2001-07-31

申请号：US08846039

申请日：1997-04-25

申请人： Vladimir Naumovich Vapnik ,  Steven Eugene Golowich

发明人： Vladimir Naumovich Vapnik ,  Steven Eugene Golowich

IPC分类号： G06E100

CPC分类号： G06F17/17

摘要： A method for estimating a real function that describes a phenomenon occurring in a space of any dimensionality is disclosed. The function is estimated by taking a series of measurements of the phenomenon being described and using those measurements to construct an expansion that has a manageable number of terms. A reduction in the number of terms is achieved by using an approximation that is defined as an expansion on kernel functions, the kernel functions forming an inner product in Hilbert space. By finding the support vectors for the measurements one specifies the expansion functions. The number of terms in an estimation according to the present invention is generally much less than the number of observations of the real world phenomenon that is being estimated. In one embodiment, the function estimation method may be used to reconstruct a radiation density image using Positron Emission Tomography (PET) scan measurements.

摘要翻译： 公开了一种用于估计描述在任何维度的空间中发生的现象的实际功能的方法。 通过对所描述的现象进行一系列测量并使用这些测量来构建具有可管理数量的扩展的功能来估计该功能。 通过使用定义为内核函数扩展的近似来实现术语数量的减少，内核函数在希尔伯特空间中形成内积。 通过查找测量的支持向量，可以指定扩展功能。 根据本发明的估计中的术语数量通常远小于正在估计的现实世界现象的观察次数。 在一个实施例中，功能估计方法可用于使用正电子发射断层扫描（PET）扫描测量来重建辐射密度图像。

公开(公告)号：US06823322B2

公开(公告)日：2004-11-23

申请号：US10077028

申请日：2002-02-14

申请人： Kim Joseph Olszewski

发明人： Kim Joseph Olszewski

IPC分类号： G06E100

CPC分类号： G06N3/0454 ,  H04W52/24

摘要： A nonlinear signal mapper that can implement any continuous one-to-one nonlinear map of baseband or intermediate-frequency digital signals. The mapping method follows a “divide-and-conquer” approach in that a nonlinear map to be implemented is piecewise decomposed into a set of simpler nonlinear component maps. The component maps are implemented using code-enabled feed-forward neural networks (FF-NNs). Each code-enabled feed-forward neural network only operates on samples of a digital input signal that lie in a specified interval of the real-valued number line. Code-enabled FF-NNs are controlled by codewords produced by a scalar quantization encoder. The quantization encoder also controls a multiplexer that directs values produced by the FF-NNs to the nonlinear mapper's output.

摘要翻译： 一种非线性信号映射器，可以实现基带或中频数字信号的任何连续的一对一非线性映射。 映射方法遵循“划分与征服”的方法，因为要实现的非线性映射被分段分解为一组更简单的非线性分量映射。 使用启用代码的前馈神经网络（FF-NN）实现分量图。 每个启用代码的前馈神经网络只对位于实数数字行的指定间隔的数字输入信号的采样进行操作。 支持代码的FF-NN由标量量化编码器产生的码字控制。 量化编码器还控制将FF-NN产生的值指向非线性映射器的输出的多路复用器。

公开(公告)号：US06643040B2

公开(公告)日：2003-11-04

申请号：US09961311

申请日：2001-09-25

申请人： Alexandre Shen ,  Fabrice Devaux

发明人： Alexandre Shen ,  Fabrice Devaux

IPC分类号： G06E100

CPC分类号： H04B10/299 ,  G02F3/022 ,  H04B2210/517

摘要： The present invention is a device for converting an RZ signal into a NRZ signal which contains an optical bistable device (5), where an output level of this device passing from a low level to a high level when an input power level crosses in an upward direction a first threshold, and returning to a low level when an input level crosses in a downward direction a second threshold below the first, the output (7) of the bistable (5) carrying the NRZ signal, and a device (2) for converting the RZ signal into a control signal of an output logic level of the optical bistable device (5) receiving the RZ signal, and delivering the control signal of the optical bistable device (5), this signal having a level above the first threshold when the RZ signal passes to 1 and which becomes lower at the second threshold only if the RZ signal passes to 0 and stays there for more than one bit time.

摘要翻译： 本发明是一种用于将RZ信号转换为包含光学双稳态器件（5）的NRZ信号的装置，其中当输入功率电平向上跨越时，该器件的输出电平从低电平传递到高电平 方向为第一阈值，并且当输入电平在向下方向上跨过第一阈值低于携带NRZ信号的双稳态（5）的输出（7）的第二阈值时返回到低电平;以及用于 将RZ信号转换为接收RZ信号的光学双稳态器件（5）的输出逻辑电平的控制信号，并且传输光学双稳态器件（5）的控制信号，该信号具有高于第一阈值的电平，当 只有当RZ信号传递到0并且停留在多于一个位的时间时，RZ信号才转到1，并且在第二阈值处变低。

公开(公告)号：US06496812B1

公开(公告)日：2002-12-17

申请号：US09571874

申请日：2000-05-13

申请人： Crystal Campaigne ,  Philip James Campaigne

发明人： Crystal Campaigne ,  Philip James Campaigne

IPC分类号： G06E100

CPC分类号： G06N3/02

摘要： A method and system for human or computer-based group-members to interact with peers to craft an action sequence to achieve a group goal. Method includes means for guiding group members on how to integrate their activities in pursuit of a specific pre-defined group goal, when given only partial understanding of how they can achieve said goal. The method identifies, selects, values and integrates group-member actions that are causal to a group achievement. The system incorporates the method along with means for recording, assigning value and reporting contributions by group members. System also includes an apparatus consisting of head-mounted microphone, voice recognition software and miniature video screen in field of view to aid data collection in applications where events occur in rapid sequence. For computer-based group members, system includes unsupervised neural network embodied in a computer mechanism and means to evaluate the instant activity and immediately relate processed information to guide the integration of group members actions.

摘要翻译： 用于人或基于计算机的组成员与对等体进行交互以制作动作序列以实现组目标的方法和系统。 方法包括指导小组成员如何仅在部分了解如何实现上述目标的情况下，指导小组成员如何整合活动来追求具体的预定义小组目标。 该方法识别，选择，评估价值，并将对群体成就产生因果关系的群组成员行为进行整合。 该系统包含方法以及组成员的记录，分配价值和报告贡献的手段。 系统还包括由头戴式麦克风，语音识别软件和视野中的微型视频屏幕组成的装置，以帮助在快速顺序发生事件的应用中的数据采集。 对于基于计算机的组成员，系统包括体现在计算机机构中的无监督神经网络，以及评估即时活动并立即将处理的信息相关联以指导组成员行为的集成的手段。

公开(公告)号：US06266656B1

公开(公告)日：2001-07-24

申请号：US09157315

申请日：1998-09-21

申请人： Kazuhiko Ohno

发明人： Kazuhiko Ohno

IPC分类号： G06E100

CPC分类号： G06F17/30707 ,  G06K9/626 ,  G06N5/025

摘要： A classification apparatus for performing effective learning type automatic classification for realistic problems of classification. The apparatus includes input unit for entering the known case data and the unknown case data, a classification ruled database for storing classification rules including the probabilistic information, a case database for storing the known case data in the form of a network based on the logical relation of conditional parts, a probability value estimating unit for estimating probability values of the results of classification using the conditional parts of the known case data and the unknown case data as entered and the rules of classification and a classification rule generating unit for evaluating the validity of the classification rules by statistic verification for suppressing generation of useless classification rules, and a negative condition searching unit for receiving all or part of the conditional parts of the known case data as entered. The classification rule generating unit has an added function of generating a classification rule including the negative condition using the negative condition searching unit.

摘要翻译： 一种分类装置，用于对实际的分类问题进行有效的学习型自动分类。 该装置包括用于输入已知病例数据和未知病例数据的输入单元，用于存储包括概率信息的分类规则的分类规则数据库，用于基于逻辑关系以网络的形式存储已知病例数据的病例数据库 条件部分的概率值估计单元，用于使用已知病例数据的条件部分和输入的未知病例数据来估计分类结果的概率值，以及用于评估有效性的分类规则和分类规则生成单元 通过用于抑制无用分类规则的生成的统计验证的分类规则，以及用于接收输入的已知病例数据的全部或部分条件部分的否定条件搜索单元。 分类规则生成单元具有使用负条件搜索单元生成包括负条件的分类规则的附加功能。

公开(公告)号：US06704718B2

公开(公告)日：2004-03-09

申请号：US09681789

申请日：2001-06-05

申请人： Chris J. C. Burges ,  Patrice Y. Simard ,  Henrique S. Malvar

发明人： Chris J. C. Burges ,  Patrice Y. Simard ,  Henrique S. Malvar

IPC分类号： G06E100

CPC分类号： H04N19/1883 ,  H04N19/105 ,  H04N19/124 ,  H04N19/187 ,  H04N19/61 ,  H04N19/63

摘要： A system and method for performing trainable nonlinear prediction of transform coefficients in data compression such that the number of bits required to represent the data is reduced. The nonlinear prediction data compression system includes a nonlinear predictor for generating predicted transform coefficients, a nonlinear prediction encoder that uses the predicted transform coefficients to encode original data, and a nonlinear prediction decoder that uses the predicted transform coefficients to decode the encoded bitstream and reconstruct the original data. The nonlinear predictor may be trained using training techniques, including a novel in-loop training technique of the present invention. The present invention also includes a method for using a nonlinear predictor to encode and decode data. The method also includes improving the performance of the nonlinear prediction data compression and decompression using several novel speedup techniques.

摘要翻译： 一种用于对数据压缩中的变换系数执行可训练非线性预测的系统和方法，使得表示数据所需的位数减少。 非线性预测数据压缩系统包括用于产生预测变换系数的非线性预测器，使用预测变换系数对原始数据进行编码的非线性预测编码器，以及使用预测变换系数对编码比特流进行解码并重建的非线性预测解码器 原始资料。 可以使用训练技术来训练非线性预测器，包括本发明的新颖的循环训练技术。 本发明还包括使用非线性预测器对数据进行编码和解码的方法。 该方法还包括使用几种新的加速技术来提高非线性预测数据压缩和解压缩的性能。

公开(公告)号：US06516306B1

公开(公告)日：2003-02-04

申请号：US09375657

申请日：1999-08-17

申请人： Rajeev Alur ,  Mihalis Yannakakis

发明人： Rajeev Alur ,  Mihalis Yannakakis

IPC分类号： G06E100

CPC分类号： G06F11/28

摘要： Model checking for message sequence charts (MSCs), message sequence chart graphs and hierarchical message sequence chart graphs (HMSCs) is provided. To verify the behavior of a given MSC, MSC graph and HMSC, a specification automaton is constructed. This specification automaton specifies the undesirable executions of the model under analysis. From the model under analysis, linearizations are defined from the model and a finite test automaton is constructed from the linearizations. The test automaton and the specification automaton are combined and it is determined whether there is an execution in the intersection. Where no state in the specification automaton is reachable from the test automaton, the model is verified.

摘要翻译： 提供消息序列图（MSCs），消息序列图表和分层消息序列图表（HMSC）的模型检查。 为了验证给定MSC，MSC图和HMSC的行为，构建了一个规范自动机。 该规范自动机规定了分析中模型的不良执行情况。 从分析模型中，从模型中定义线性化，并从线性化构建有限测试自动机。 测试自动机和规范自动机组合起来，确定交叉路口是否有执行。 在规范自动机中无法从测试自动机到达的状态下，验证模型。

公开(公告)号：US06292280B1

公开(公告)日：2001-09-18

申请号：US09096387

申请日：1998-06-12

申请人： Pierpaolo Boffi ,  Mario Martinelli ,  Diego Mottarella ,  Davide Piccinin

发明人： Pierpaolo Boffi ,  Mario Martinelli ,  Diego Mottarella ,  Davide Piccinin

IPC分类号： G06E100

CPC分类号： H04B10/11 ,  H04B10/1141 ,  H04Q11/0003 ,  H04Q2011/0015 ,  H04Q2011/002 ,  H04Q2011/0024 ,  H04Q2011/0026 ,  H04Q2011/0035 ,  H04Q2011/0039 ,  H04Q2011/0041

摘要： An optical device is provided for processing a digital optical signal in parallel and in free space. A device consistent with the invention provides for inputting a digital optical signal in guided propagation, converting the digital optical signal to n digital optical signals in guided propagation, and converting the n digital optical signals to n digital optical signals in free space. In addition, the invention provides for selecting a bit preselected from the at least one temporal series of n bits of each of the n digital optical signals, modifying optically, at least one bit of the spatial figure of the n bits, and outputting the at least one bit of the spatial figure of n bits.

摘要翻译： 提供一种用于并行和自由空间处理数字光信号的光学装置。 符合本发明的装置提供了在导向传播中输入数字光信号，将数字光信号转换成n导数传播中的数字光信号，并将n个数字光信号转换成自由空间中的n个数字光信号。 另外，本发明提供从n个数字光信号中的每一个n帧的至少一个时间序列预先选择的位，光学修改n位空间图的至少一位，并输出在 n位的空间图的至少一位。