知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

16 records (took 0.01 seconds)

    DELAYED ACTIONS FOR A DECENTRALIZED SYSTEM OF LEARNING DEVICES
    3.
    发明申请
    DELAYED ACTIONS FOR A DECENTRALIZED SYSTEM OF LEARNING DEVICES 有权
    延迟学习设备分散系统的行动

    公开(公告)号：US20140351374A1

    公开(公告)日：2014-11-27

    申请号：US14286476

    申请日：2014-05-23

    Applicant: QUALCOMM Incorporated

    Inventor： Michael-David Nakayoshi CANOY

    IPC: H04L29/08

    CPC classification number: H04L67/10 ,  G06N99/005

    Abstract: An embodiment delay device for use within a decentralized system of learning device delays broadcast messages to introduce a time shift into events. The delay device may receive a first message from a triggering device, generate a first pattern using at least a first event based on the received first message, determine whether the first pattern matches a known trigger pattern, wait a predetermined delay period in response to determining that the first pattern matches the known trigger pattern, and broadcast a second message in response to the predetermined delay period expiring. Delay periods may be user-configurable, such as via user inputs (e.g., dials, sliders, etc.) or learned based on messages from responding devices. The second message may be similar to the first message or a distinct message indicating the elapse of the delay period.

    Abstract translation: 在分散式学习设备系统内使用的实施例延迟装置延迟广播消息以将时移引入事件。 所述延迟装置可以从触发装置接收第一消息，基于所接收的第一消息，使用至少第一事件生成第一模式，确定所述第一模式是否与已知触发模式匹配，响应于确定，等待预定的延迟周期 第一模式与已知的触发模式匹配，并且响应于预定的延迟时段到期而广播第二消息。 延迟周期可以是用户可配置的，例如通过用户输入（例如，拨号盘，滑块等）或基于来自响应设备的消息学习。 第二消息可以类似于第一消息或指示延迟时间段经过的明确消息。

    MONITORING NEURAL NETWORKS WITH SHADOW NETWORKS
    5.
    发明申请
    MONITORING NEURAL NETWORKS WITH SHADOW NETWORKS 有权
    用阴影网络监测神经网络

    公开(公告)号：US20150206049A1

    公开(公告)日：2015-07-23

    申请号：US14162646

    申请日：2014-01-23

    Applicant: QUALCOMM Incorporated

    Inventor： Michael-David Nakayoshi CANOY ,  Paul BENDER

    IPC: G06N3/08

    CPC classification number: G06N3/0454 ,  G06N3/049

    Abstract: A method for generating an event includes monitoring a first neural network with a second neural network. The method also includes generating an event based at least in part on the monitoring. The event is generated at the second neural network.

    Abstract translation: 一种用于产生事件的方法包括利用第二神经网络来监测第一神经网络。 该方法还包括至少部分地基于监视来生成事件。 该事件在第二神经网络中产生。

    IMPLEMENTING STRUCTURAL PLASTICITY IN AN ARTIFICIAL NERVOUS SYSTEM
    6.
    发明申请
    IMPLEMENTING STRUCTURAL PLASTICITY IN AN ARTIFICIAL NERVOUS SYSTEM 有权
    在人造神经系统中实施结构塑性

    公开(公告)号：US20150081607A1

    公开(公告)日：2015-03-19

    申请号：US14157143

    申请日：2014-01-16

    Applicant: QUALCOMM Incorporated

    Inventor： Jason Frank HUNZINGER ,  Michael-David Nakayoshi CANOY ,  Paul Edward BENDER ,  Victor Hokkiu CHAN ,  Gina Marcela ESCOBAR MORA

    IPC: G06N3/04 ,  G06N3/08

    CPC classification number: G06N3/04 ,  G06N3/049 ,  G06N3/08 ,  G06N3/082

    Abstract: Methods and apparatus are provided for implementing structural plasticity in an artificial nervous system. One example method for altering a structure of an artificial nervous system generally includes determining a synapse in the artificial nervous system for reassignment, determining a first artificial neuron and a second artificial neuron for connecting via the synapse, and reassigning the synapse to connect the first artificial neuron with the second artificial neuron. Another example method for operating an artificial nervous system, generally includes determining a synapse in the artificial nervous system for assignment; determining a first artificial neuron and a second artificial neuron for connecting via the synapse, wherein at least one of the synapse or the first and second artificial neurons are determined randomly or pseudo-randomly; and assigning the synapse to connect the first artificial neuron with the second artificial neuron.

    Abstract translation: 提供了在人造神经系统中实现结构可塑性的方法和装置。 用于改变人造神经系统的结构的一个示例性方法通常包括确定用于重新分配的人造神经系统中的突触，确定第一人造神经元和第二人造神经元以经由突触连接，以及重新分配突触以连接第一人工神经元 神经元与第二个人造神经元。 用于操作人造神经系统的另一示例性方法通常包括确定用于分配的人造神经系统中的突触; 确定经由所述突触连接的第一人造神经元和第二人造神经元，其中所述突触或所述第一和第二人造神经元中的至少一个被随机地或伪随机地确定; 并分配突触将第一人工神经元与第二人造神经元连接。

    METHODS AND APPARATUS FOR MODULATING THE TRAINING OF A NEURAL DEVICE
    7.
    发明申请
    METHODS AND APPARATUS FOR MODULATING THE TRAINING OF A NEURAL DEVICE 有权
    调制神经器械训练的方法和装置

    公开(公告)号：US20150052093A1

    公开(公告)日：2015-02-19

    申请号：US14079181

    申请日：2013-11-13

    Applicant: QUALCOMM Incorporated

    Inventor： Michael-David Nakayoshi CANOY ,  Yinyin LIU ,  Anthony SARAH ,  Adrienne MILNER

    IPC: G06N3/08

    CPC classification number: G06N3/08 ,  G06N3/049

    Abstract: Methods and apparatus are provided for training a neural device having an artificial nervous system by modulating at least one training parameter during the training. One example method for training a neural device having an artificial nervous system generally includes observing the neural device in a training environment and modulating at least one training parameter based at least in part on the observing. For example, the training apparatus described herein may modify the neural device's internal learning mechanisms (e.g., spike rate, learning rate, neuromodulators, sensor sensitivity, etc.) and/or the training environment's stimuli (e.g., move a flame closer to the device, make the scene darker, etc.). In this manner, the speed with which the neural device is trained (i.e., the training rate) may be significantly increased compared to conventional neural device training systems.

    Abstract translation: 提供了用于通过在训练期间调制至少一个训练参数来训练具有人造神经系统的神经装置的方法和装置。 用于训练具有人造神经系统的神经装置的一个示例性方法通常包括在训练环境中观察神经装置并且至少部分地基于观察来调制至少一个训练参数。 例如，本文描述的训练装置可以修改神经装置的内部学习机制（例如，尖峰率，学习速率，神经调节器，传感器灵敏度等）和/或训练环境的刺激（例如，将火焰移动到设备附近 ，使场景更暗等）。 以这种方式，与传统的神经元装置训练系统相比，神经装置训练的速度（即，训练速率）可以显着增加。

    NEURAL WATCHDOG
    8.
    发明申请
    NEURAL WATCHDOG 有权
    神经手表

    公开(公告)号：US20150178617A1

    公开(公告)日：2015-06-25

    申请号：US14139732

    申请日：2013-12-23

    Applicant: QUALCOMM Incorporated

    Inventor： Michael-David Nakayoshi CANOY ,  Jason Frank HUNZINGER

    IPC: G06N3/02

    CPC classification number: G06N3/049

    Abstract: A method of monitoring a neural network includes monitoring activity of the neural network. The method also includes detecting a condition based on the activity. The method further includes performing an exception event based on the detected condition.

    Abstract translation: 监测神经网络的方法包括监测神经网络的活动。 该方法还包括基于活动来检测条件。 该方法还包括基于检测到的条件执行异常事件。

    Mobile Coprocessor System and Methods
    9.
    发明申请
    Mobile Coprocessor System and Methods 有权
    移动协处理器系统和方法

    公开(公告)号：US20150106823A1

    公开(公告)日：2015-04-16

    申请号：US14053790

    申请日：2013-10-15

    Applicant: QUALCOMM Incorporated

    Inventor： Michael-David Nakayoshi CANOY ,  Sisil Mehta ,  Kirtika Ruchandani

    IPC: G06F9/50

    CPC classification number: G06F9/505 ,  G06F9/5044 ,  G06F2209/509 ,  Y02D10/22

    Abstract: Embodiments include apparatuses, systems, and methods mobile coprocessing. A connection is established between a mobile device and an auxiliary computing device. The mobile device implements a CPU abstraction layer and a virtual CPU between a software stack and a CPU of the mobile device. The abstraction layer allows for the mobile device to offload tasks to the auxiliary computing device while the software stack interacts with the abstraction layer as if the tasks are being executed by the CPU of the mobile device. The mobile device of allocates tasks to the auxiliary computing device based on various parameters, including properties of the auxiliary computing device, metrics of the connection, and priorities of the tasks.

    Abstract translation: 实施例包括移动协处理的装置，系统和方法。 在移动设备和辅助计算设备之间建立连接。 移动设备在软件栈和移动设备的CPU之间实现CPU抽象层和虚拟CPU。 抽象层允许移动设备将任务卸载到辅助计算设备，同时软件堆栈与抽象层进行交互，就好像该任务正由移动设备的CPU执行。 移动设备基于各种参数（包括辅助计算设备的属性，连接的度量以及任务的优先级）来将辅助计算设备的任务分配给辅助计算设备。

    REQUESTING PROXIMATE RESOURCES BY LEARNING DEVICES
    10.
    发明申请
    REQUESTING PROXIMATE RESOURCES BY LEARNING DEVICES 审中-公开
    通过学习设备要求临时资源

    公开(公告)号：US20140351181A1

    公开(公告)日：2014-11-27

    申请号：US14286362

    申请日：2014-05-23

    Applicant: QUALCOMM Incorporated

    Inventor： Michael-David Nakayoshi CANOY ,  Anne Katrin Konertz ,  Kiet Chau ,  Kenchin Lai

    IPC: G06N99/00 ,  G06N5/04

    CPC classification number: G06N20/00 ,  G06N5/04

    Abstract: Various embodiments for a learning device to improve the performance of learned behaviors by requesting information from proximate devices within a decentralized system including a learning device method for generating, by the learning device, a first pattern based upon one or more obtained events, determining whether the first pattern exactly matches a known second pattern, determining whether the first pattern matches the second pattern within a predefined threshold in response to determining that the first pattern does not exactly match the second pattern, identifying a missing event of the second pattern in response to determining that the first pattern matches the second pattern within the predefined threshold, and broadcasting, by the learning device, a message requesting data related to the identified missing event. Data received in response to request messages may be used to recognize that the known second pattern is matched.

    Abstract translation: 用于学习设备的各种实施例，通过从分散系统内的近端设备请求信息来提高学习行为的性能，包括用于由学习设备基于一个或多个获得的事件生成第一模式的学习设备方法， 第一模式与已知的第二模式精确匹配，响应于确定第一模式与第二模式不完全匹配，确定第一模式是否与预定阈值内的第二模式相匹配，以响应确定第二模式识别第二模式的丢失事件 第一模式与预定阈值内的第二模式相匹配，并且由学习装置广播请求与所识别的丢失事件有关的数据的消息。 可以使用响应于请求消息接收到的数据来识别已知的第二模式是匹配的。

Patent Agency Ranking