发明名称 |
HYPER-PARAMETER SELECTION FOR DEEP CONVOLUTIONAL NETWORKS |
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. |
申请公布号 |
US2016224903(A1) |
申请公布日期 |
2016.08.04 |
申请号 |
US201514848296 |
申请日期 |
2015.09.08 |
申请人 |
QUALCOMM Incorporated |
发明人 |
TALATHI Sachin Subhash;JULIAN David Jonathan |
分类号 |
G06N99/00;G06N3/08 |
主分类号 |
G06N99/00 |
代理机构 |
|
代理人 |
|
主权项 |
1. A method of selecting hyper-parameters for training a deep convolutional network, comprising:
selecting a number of network architectures as part of a database, each of the network architectures including at least one local logistic regression layer; training each of the network architectures to generate a corresponding validation error that is stored in the database; estimating a threshold error for identifying a good set of network architectures and a bad set of network architectures based at least in part on validation errors in the database; choosing a next potential hyper-parameter, corresponding to a next network architecture, based at least in part on a metric that is a function of the good set of network architectures; and selecting a network architecture, from among a plurality of next network architectures, with a lowest validation error. |
地址 |
San Diego CA US |