专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20160224903A1 HYPER-PARAMETER SELECTION FOR DEEP CONVOLUTIONAL NETWORKS 审中-公开
标题翻译：深层调节网络的参数选择
公开(公告)号：US20160224903A1
公开(公告)日：2016-08-04
申请号：US14848296
申请日：2015-09-08
申请人： QUALCOMM Incorporated
发明人： Sachin Subhash TALATHI , David Jonathan JULIAN
IPC分类号： G06N99/00 , G06N3/08
CPC分类号： G06N99/005 , G06N3/08 , G06N3/082 , G06N7/005
摘要： Hyper-parameters are selected for training a deep convolutional network by selecting a number of network architectures as part of a database. Each of the network architectures includes one or more local logistic regression layer and is trained to generate a corresponding validation error that is stored in the database. A threshold error for identifying a good set of network architectures and a bad set of network architectures may be estimated based on validation errors in the database. The method also includes choosing a next potential hyper-parameter, corresponding to a next network architecture, based on a metric that is a function of the good set of network architectures. The method further includes selecting a network architecture, from among next network architectures, with a lowest validation error.
摘要翻译：选择超参数以通过选择多个网络架构作为数据库的一部分来训练深卷积网络。每个网络架构包括一个或多个本地逻辑回归层，并被训练以产生存储在数据库中的对应验证错误。可以基于数据库中的验证错误来估计用于识别良好的一组网络架构和一组坏的网络架构的阈值误差。该方法还包括基于作为良好网络体系结构的函数的度量来选择对应于下一个网络体系结构的下一个潜在的超参数。该方法还包括从下一个网络体系结构中选择具有最低验证错误的网络架构。

2. 发明申请

US20150324689A1 CUSTOMIZED CLASSIFIER OVER COMMON FEATURES 审中-公开
标题翻译：自定义分类器通用功能
公开(公告)号：US20150324689A1
公开(公告)日：2015-11-12
申请号：US14483075
申请日：2014-09-10
申请人： QUALCOMM Incorporated
发明人： Casimir Matthew WIERZYNSKI , David Jonathan JULIAN , Anthony SARAH
IPC分类号： G06N3/08
CPC分类号： G06N3/049 , G06N3/0454 , G06N3/08
摘要： A method of updating a set of classifiers includes applying a first set of classifiers to a first set of data. The method further includes requesting, from a remote device, a classifier update based on an output of the first set of classifiers or a performance measure of the application of the first set of classifiers.
摘要翻译：更新一组分类器的方法包括将第一组分类器应用于第一组数据。该方法还包括从远程设备请求基于第一组分类器的输出的分类器更新或第一组分类器的应用的性能测量。

3. 发明申请

US20150269481A1 DIFFERENTIAL ENCODING IN NEURAL NETWORKS 审中-公开
标题翻译：神经网络中的差分编码
公开(公告)号：US20150269481A1
公开(公告)日：2015-09-24
申请号：US14513155
申请日：2014-10-13
申请人： QUALCOMM Incorporated
发明人： Venkata Sreekanta Reddy ANNAPUREDDY , David Jonathan JULIAN , Regan Blythe TOWAL , Yinyin LIU
IPC分类号： G06N3/08 , G06N3/04
CPC分类号： G06N3/0445 , G06N3/0481 , G06N3/049
摘要： Differential encoding in a neural network includes predicting an activation value for a neuron in the neural network based on at least one previous activation value for the neuron. The encoding further includes encoding a value based on a difference between the predicted activation value and an actual activation value for the neuron in the neural network.
摘要翻译：神经网络中的差分编码包括基于神经元的至少一个先前激活值来预测神经网络中神经元的激活值。编码还包括基于预测的激活值和神经网络中的神经元的实际激活值之间的差来编码值。

4. 发明申请

US20150206050A1 CONFIGURING NEURAL NETWORK FOR LOW SPIKING RATE 审中-公开
标题翻译：配置低速扫描速率的神经网络
公开(公告)号：US20150206050A1
公开(公告)日：2015-07-23
申请号：US14449092
申请日：2014-07-31
申请人： QUALCOMM Incorporated
发明人： Sachin Subhash TALATHI , David Jonathan JULIAN , Venkata Sreekanta Reddy ANNAPUREDDY
IPC分类号： G06N3/08 , G06N3/04
CPC分类号： G06N3/08 , G06N3/0481 , G06N3/049 , G06N3/082
摘要： A method for selecting a neuron model with a user defined firing rate for operating in a neural network includes selecting the neuron model based on a selected firing rate bandwidth.
摘要翻译：用于在神经网络中操作的用户定义的发射速率来选择神经元模型的方法包括基于所选择的发射速率带宽选择神经元模型。

5. 发明申请

US20150317557A1 TEMPORAL SPIKE ENCODING FOR TEMPORAL LEARNING 审中-公开
标题翻译：用于时间学习的TEMPORAL SPIKE编码
公开(公告)号：US20150317557A1
公开(公告)日：2015-11-05
申请号：US14315531
申请日：2014-06-26
申请人： QUALCOMM INCORPORATED
发明人： David Jonathan JULIAN , Venkata Sreekanth Reddy ANNAPUREDDY , Kristopher David PETERSON
IPC分类号： G06N3/08
CPC分类号： G06N3/049 , G06N3/08
摘要： Certain aspects of the present disclosure support methods and apparatus for temporal spike encoding for temporal learning in an artificial nervous system. The temporal spike encoding for temporal learning can comprise obtaining sensor data being input into the artificial nervous system, processing the sensor data to generate feature vectors, converting element values of the feature vectors into delays, and causing at least one artificial neuron of the artificial nervous system to spike at times based on the delays.
摘要翻译：本公开的某些方面支持用于人造神经系统中的时间学习的时间尖峰编码的方法和装置。用于时间学习的时间尖峰编码可以包括获得输入到人造神经系统的传感器数据，处理传感器数据以产生特征向量，将特征向量的元素值转换为延迟，并且引起人造神经的至少一个人造神经元系统根据延迟时间飙升。

6. 发明申请

US20150269479A1 CONVERSION OF NEURON TYPES TO HARDWARE 审中-公开
标题翻译：神经类型转换为硬件
公开(公告)号：US20150269479A1
公开(公告)日：2015-09-24
申请号：US14286277
申请日：2014-05-23
申请人： QUALCOMM Incorporated
发明人： David Jonathan JULIAN , Jan Krzysztof WEGRZYN
IPC分类号： G06N3/08
CPC分类号： G06N3/0481
摘要： Certain aspects of the present disclosure support a method and apparatus for conversion of neuron types to a hardware implementation of an artificial nervous system. According to certain aspects, at least one of synapse weights of the artificial nervous system, neuron input channel resistances associated with a neuron model for neuron instances of the artificial nervous system, or neuron input channel potentials associated with the neuron model can be normalized by one or more factors. A linear transformation can be determined for mapping of parameters of the neuron model. Then, the linear transformation can be applied to the parameters of the neuron model to obtain transformed parameters of the neuron model, and at least one of inputs to the neuron instances or dynamics of the neuron model based may be updated based at least in part on the transformed parameters.
摘要翻译：本公开的某些方面支持用于将神经元类型转换为人造神经系统的硬件实现的方法和装置。根据某些方面，人造神经系统的突触重量中的至少一个，与人造神经系统的神经元实例的神经元模型相关联的神经元输入通道电阻或与神经元模型相关联的神经元输入通道电位可以被一个或更多因素。可以确定神经元模型的参数映射的线性变换。然后，线性变换可以应用于神经元模型的参数以获得神经元模型的变换参数，并且至少基于神经元模型的神经元实例或动力学的输入中的至少一个可以至少部分地基于变换参数。

7. 发明申请

US20150248609A1 NEURAL NETWORK ADAPTATION TO CURRENT COMPUTATIONAL RESOURCES 审中-公开
标题翻译：神经网络适应当前的计算资源
公开(公告)号：US20150248609A1
公开(公告)日：2015-09-03
申请号：US14268372
申请日：2014-05-02
申请人： QUALCOMM INCORPORATED
发明人： David Jonathan JULIAN , Ilwoo CHANG
IPC分类号： G06N3/08 , G06N3/06
CPC分类号： G06N3/082 , G06N3/0481
摘要： Methods and apparatus are provided for processing in an artificial nervous system. According to certain aspects, resolution of one or more functions performed by processing units of a neuron model may be reduced, based at least in part on availability of computational resources or a power target or budget. The reduction in resolution may be compensated for by adjusting one or more network weights.
摘要翻译：提供了用于在人造神经系统中加工的方法和装置。根据某些方面，可以至少部分地基于计算资源的可用性或功率目标或预算来减少由神经元模型的处理单元执行的一个或多个功能的分辨率。可以通过调整一个或多个网络权重来补偿分辨率的降低。

8. 发明申请

US20170091619A1 SELECTIVE BACKPROPAGATION 审中-公开
公开(公告)号：US20170091619A1
公开(公告)日：2017-03-30
申请号：US15081780
申请日：2016-03-25
申请人： QUALCOMM Incorporated
发明人： Regan Blythe TOWAL , David Jonathan JULIAN
IPC分类号： G06N3/08 , G06N3/04
CPC分类号： G06N3/084 , G06K9/4628 , G06N3/0472
摘要： The balance of training data between classes for a machine learning model is modified. Adjustments are made at the gradient stage where selective backpropagation is utilized to modify a cost function to adjust or selectively apply the gradient based on the class example frequency in the data sets. The factor for modifying the gradient may be determined based on a ratio of the number of examples of the class with a fewest members to the number of examples of a present class. The gradient associated with the present class is modified based on the above determined factor.

9. 发明申请

US20150206048A1 CONFIGURING SPARSE NEURONAL NETWORKS 审中-公开
标题翻译：配置稀疏神经网络
公开(公告)号：US20150206048A1
公开(公告)日：2015-07-23
申请号：US14449101
申请日：2014-07-31
申请人： QUALCOMM Incorporated
发明人： Sachin Subhash TALATHI , David Jonathan JULIAN , Venkata Sreekanta Reddy ANNAPUREDDY
IPC分类号： G06N3/04
CPC分类号： G06N3/08 , G06N3/0481 , G06N3/049 , G06N3/082
摘要： A method for selecting a reduced number of model neurons in a neural network includes generating a first sparse set of non-zero decoding vectors. Each of the decoding vector is associated with a synapse between a first neuron layer and a second neuron layer. The method further includes implementing the neural network only with selected model neurons in the first neuron layer associated with the non-zero decoding vectors.
摘要翻译：用于在神经网络中选择减少数量的模型神经元的方法包括生成非零解码矢量的第一稀疏集合。每个解码矢量与第一神经元层和第二神经元层之间的突触相关联。该方法还包括仅在与非零解码矢量相关联的第一神经元层中实施神经网络与选定的模型神经元。

10. 发明申请

US20140351186A1 SPIKE TIME WINDOWING FOR IMPLEMENTING SPIKE-TIMING DEPENDENT PLASTICITY (STDP) 审中-公开
标题翻译： SPIKE时间窗口执行SPIKE-TIMING相关塑料（STDP）
公开(公告)号：US20140351186A1
公开(公告)日：2014-11-27
申请号：US14084302
申请日：2013-11-19
申请人： QUALCOMM Incorporated
发明人： David Jonathan JULIAN , Jeffrey Alexander LEVIN , Kristopher PETERSON , Ravindra Manohar PATWARDHAN
IPC分类号： G06N3/08
CPC分类号： G06N3/08 , G06N3/049 , G06N3/10
摘要： Methods and apparatus are provided for implementing spike-timing dependent plasticity (STDP) using windowing of spikes. One example method for operating an artificial nervous system generally includes recording spike times for a first artificial neuron, recording spike times for a second artificial neuron coupled to the first artificial neuron via a synapse, processing spikes for the second artificial neuron according to a window based at least in part on the spike times for the first artificial neuron, and updating a parameter (e.g., a weight or a delay) of the synapse based on the processing.
摘要翻译：提供了使用尖峰窗口来实现尖峰时序相关可塑性（STDP）的方法和装置。用于操作人造神经系统的一个示例性方法通常包括记录第一人造神经元的尖峰时间，通过突触记录耦合到第一人造神经元的第二人造神经元的尖峰时间，根据基于窗口的第二人造神经元的尖峰至少部分地基于第一人造神经元的尖峰时间，以及基于该处理更新突触的参数（例如，重量或延迟）。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式