基本信息:
- 专利标题: DEEP NEURAL NETWORK PROCESSOR WITH INTERLEAVED BACKPROPAGATION
- 申请号:PCT/US2018/060611 申请日:2018-11-13
- 公开(公告)号:WO2019094882A1 公开(公告)日:2019-05-16
- 发明人: GOULDING, John R. , MIXTER, John E. , MUCHA, David R. , GANGWER, Troy A. , SILVA, Ryan D.
- 申请人: RAYTHEON COMPANY
- 申请人地址: 870 Winter Street Waltham, Massachusetts 02451-1449 US
- 专利权人: RAYTHEON COMPANY
- 当前专利权人: RAYTHEON COMPANY
- 当前专利权人地址: 870 Winter Street Waltham, Massachusetts 02451-1449 US
- 代理机构: PERDOK, Monique M. et al.
- 优先权: US15/810,946 20171113
- 主分类号: G06N3/063
- IPC分类号: G06N3/063 ; G06N3/04 ; G06N3/08
摘要:
Processing circuitry for a deep neural network can include input/output ports, and a plurality of neural network layers coupled in order from a first layer to a last layer, each of the plurality of neural network layers including a plurality of weighted computational units having circuitry to interleave forward propagation of computational unit input values from the first layer to the last layer and backward propagation of output error values from the last layer to the first layer.
IPC结构图谱:
G | 物理 |
--G06 | 计算;推算;计数 |
----G06N | 基于特定计算模型的计算机系统 |
------G06N3/00 | 基于生物学模型的计算机系统 |
--------G06N3/02 | .采用神经网络模型 |
----------G06N3/06 | ..物理实现,即神经网络、神经元或神经元部分的硬件实现 |
------------G06N3/063 | ...采用电的 |