会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明申请
    • ANALOG ACCUMULATOR
    • 模拟累加器
    • US20140292375A1
    • 2014-10-02
    • US13853870
    • 2013-03-29
    • STMicroelectronics Asia Pacific Pte. Ltd.STMicroelectronics S.r.l.
    • Paolo AngeliniYannick GuedonMing Zi Zhu
    • G11C27/02
    • G06F3/044G03G7/00G06F3/0418G06G7/00
    • Accumulators that operate to fully or partially remove noise from a signal, including removing noise inserted into the signal by the accumulator itself. In some embodiments, an accumulator may be operated in a sampling phase and a transfer phase each time the accumulator samples an input signal. In some such embodiments, an op-amp of an accumulation circuit of the accumulator may be auto-zeroed during some or all of the sampling phases of an accumulation period. In some embodiments in which the op-amp is auto-zeroed during some or all of the sampling phases, the accumulation circuit may include a holding capacitor that, during an auto-zeroing process, holds a value output by the op-amp during a prior transfer phase. Including such a holding capacitor in an accumulator may reduce a voltage that the op-amp output rises following the auto-zero process, which may reduce a bandwidth and noise of the accumulation circuit.
    • 用于完全或部分地从信号中去除噪声的累加器,包括消除蓄电池本身插入到信号中的噪声。 在一些实施例中,每当累加器对输入信号进行采样时,累加器可以在采样阶段和传送阶段中操作。 在一些这样的实施例中,累加器的累加电路的运算放大器可以在累积周期的某些或全部采样阶段期间自动归零。 在一些实施例中,在某些或所有采样相位期间运算放大器自动归零,其中累积电路可以包括保持电容器,其在自动归零过程期间保持由运算放大器输出的值 先前转移阶段。 在存储器中包括这种保持电容器可以降低运算放大器输出在自动归零过程之后上升的电压,这可以降低积累电路的带宽和噪声。
    • 2. 发明授权
    • Application of hebbian and anti-hebbian learning to nanotechnology-based physical neural networks
    • hebbian和anti-hebbian学习在纳米技术的物理神经网络的应用
    • US07412428B2
    • 2008-08-12
    • US10748631
    • 2003-12-30
    • Alex Nugent
    • Alex Nugent
    • G06E1/00
    • G06N3/063B82Y10/00G06E1/00G06E3/00G06G7/00G06N3/04G06N3/08G06N99/007Y10S977/70Y10S977/712Y10S977/742
    • Methods and systems are disclosed herein in which a physical neural network can be configured utilizing nanotechnology. Such a physical neural network can comprise a plurality of molecular conductors (e.g., nanoconductors) which form neural connections between pre-synaptic and post-synaptic components of the physical neural network. Additionally, a learning mechanism can be applied for implementing Hebbian learning via the physical neural network. Such a learning mechanism can utilize a voltage gradient or voltage gradient dependencies to implement Hebbian and/or anti-Hebbian plasticity within the physical neural network. The learning mechanism can also utilize pre-synaptic and post-synaptic frequencies to provide Hebbian and/or anti-Hebbian learning within the physical neural network.
    • 本文公开的方法和系统,其中可以使用纳米技术来配置物理神经网络。 这样的物理神经网络可以包括形成物理神经网络的突触前和突触前组件之间的神经连接的多个分子导体(例如,纳米电感器)。 另外,学习机制可以通过物理神经网络实现Hebbian学习。 这种学习机制可以利用电压梯度或电压梯度依赖性在物理神经网络内实现Hebbian和/或抗Hebbian可塑性。 学习机制还可以利用突触前和突触后频率在物理神经网络内提供Hebbian和/或反Hebbian学习。
    • 3. 发明授权
    • Output control device for electrical apparatus
    • 电气设备输出控制装置
    • US4704677A
    • 1987-11-03
    • US860385
    • 1986-05-06
    • Hironobu NagashimaNoriyuki Asai
    • Hironobu NagashimaNoriyuki Asai
    • G05B11/36G05B11/01G06G7/00
    • G05B11/01G05B11/36G06G7/00
    • An output control device for controlling the output of an electrical apparatus such as a laser in which a desired accuracy of control is maintained irrespective of the magnitude of the target value of the output. A first multiplication circuit multiplies the output indicating signal by a value which is dependent upon the reciprocal of the target value. The output of the first multiplication circuit is compared with a predetermined reference value to form a feedback signal having a component applied to control the output in such a manner that the absolute value of the difference between the output of the first multiplication circuit and the reference value tends to be decreased. A second multiplication circuit multiplies the comparison value output by the reciprocal of the output of the first multiplication circuit, and the resulting signal is applied to control the output of the electrical apparatus.
    • 一种输出控制装置,用于控制诸如激光器的电气设备的输出,其中保持期望的控制精度,而与输出的目标值的大小无关。 第一乘法电路将输出指示信号乘以取决于目标值的倒数的值。 将第一乘法电路的输出与预定的参考值进行比较,以形成具有应用分量的反馈信号,以便以这样的方式控制输出,使得第一乘法电路的输出与参考值之间的差的绝对值 往往会减少。 第二乘法电路将比较值输出乘以第一乘法电路的输出的倒数,并施加所得到的信号以控制电气设备的输出。
    • 4. 发明申请
    • ANALOG ARITHMETIC CIRCUIT, SEMICONDUCTOR DEVICE, AND ELECTRONIC DEVICE
    • 模拟算术电路,半导体器件和电子器件
    • US20170017285A1
    • 2017-01-19
    • US15124543
    • 2015-03-02
    • Semiconductor Energy Laboratory Co., Ltd.
    • Yoshiyuki KUROKAWATakayuki IKEDAShunpei YAMAZAKI
    • G06F1/32
    • G06F1/32G06F1/3287G06G7/00Y02D10/171
    • The power consumption of an analog arithmetic circuit is reduced. The analog arithmetic circuit includes a plurality of first circuits. An output terminal of the k-th (k is a natural number) first circuit is connected to an input terminal of the k+1-th first circuit. Each of the first circuits includes a memory circuit which holds an analog signal, a second circuit which performs arithmetic processing using the analog signal, a switch which controls power supply to the second circuit, and a controller. The conduction state of the switch included in the k-th first circuit is controlled by the controller included in the k+1-th first circuit. The arithmetic processing performed by the second circuit included in the k+1-th first circuit is started by the controller included in the k+1-th first circuit.
    • 模拟运算电路的功耗降低。 模拟运算电路包括多个第一电路。 第k(k是自然数)的第一电路的输出端连接到第k + 1个第一电路的输入端。 每个第一电路包括保持模拟信号的存储电路,使用模拟信号进行算术处理的第二电路,控制对第二电路的电源的开关和控制器。 包括在第k个第一电路中的开关的导通状态由包括在第k + 1个第一电路中的控制器控制。 包括在第k + 1个第一电路中的第二电路进行的运算处理由包括在第k + 1个第一电路中的控制器启动。
    • 8. 发明申请
    • Method for quantifying amplitude of a response of a biological network
    • 量化生物网络响应幅度的方法
    • US20120030162A1
    • 2012-02-02
    • US13149022
    • 2011-05-31
    • Ty Matthew ThomsonDexter Roydon PrattWilliam M. Ladd
    • Ty Matthew ThomsonDexter Roydon PrattWilliam M. Ladd
    • G06N5/02
    • G06N3/126G06E1/00G06E3/00G06F15/18G06F19/12G06G7/00G06N3/08G06N99/005
    • One or more measurement signatures are derived from a knowledge base of casual biological facts, where a signature is a collection of measured node entities and their expected directions of change with respect to a reference node. The knowledge base may be a directed network of experimentally-observed casual relationships among biological entities and processes, and a reference node represents a perturbation. A degree of activation of a signature is then assessed by scoring one or more “differential” data sets against the signature to compute an amplitude score. The amplitude score quantifies fold-changes of measurements in the signature. In one particular embodiment, the amplitude score is a weighted average of adjusted log-fold changes of measured node entities in the signature, wherein an adjustment applied to the log-fold changes is based on their expected direction of change. In an alternative embodiment, the amplitude score is based on quantity effects.
    • 一个或多个测量签名来源于休闲生物事实的知识库,其中签名是测量节点实体的集合及其相对于参考节点的预期改变方向。 知识库可以是生物实体和过程之间的实验观察到的随机关系的定向网络,参考节点表示扰动。 然后通过对一个或多个相对于签名的“差分”数据集来评估签名的激活程度以计算振幅得分。 幅度分数量化签名中测量的折叠变化。 在一个具体实施例中,幅度得分是签名中测量节点实体的调整的对数倍变化的加权平均值,其中应用于对数倍变化的调整基于其预期的改变方向。 在替代实施例中,幅度得分基于数量效应。