Computer system and computation method using recurrent neural network
A recurrent neural network, computer system technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve the problems of complex recurrent neural network learning process and high computational cost
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0090] figure 1 It is a diagram showing a configuration example of the computer 100 for realizing the reserve pool calculation of the first embodiment.
[0091] The computer 100 has a computing device 101 , a memory 102 and a network interface 103 .
[0092] The computing device 101 executes processing according to a program. A processor, an FPGA (Field Programmable Gate Array, Field Programmable Gate Array) and the like can be considered as the computing device 101 . Predetermined functional units are realized by the arithmetic unit 101 executing processing according to the program. In the following description, when the processing is described using a functional unit as a subject, it means that the computing device 101 executes a program that realizes the functional unit.
[0093] The memory 102 stores programs executed by the computing device 101 and information used by the programs. Furthermore, the memory 102 includes a work area temporarily used by programs.
[009...
Embodiment 2
[0166]In Embodiment 1, the input unit 111 , the reservoir unit 112 , and the output unit 113 are realized as software, whereas in Embodiment 2, they are realized using hardware. Hereinafter, details of Embodiment 2 will be described.
[0167] The nonlinear node 200 of the reservoir unit 112 can be realized using hardware such as electronic circuits and optical elements. As an electronic circuit, a Macky-Glass circuit and a source-drain current of a MOSFET can be used. As the optical element, an MZ interferometer and an optical waveguide showing nonlinear characteristics such as saturated absorption can be used.
[0168] In Embodiment 2, a computer in which the reserve tank unit 112 is realized using an optical waveguide will be described.
[0169] Optical devices have high-speed communication and network characteristics with low propagation loss in optical waveguides, so they can be flexibly applied to high-speed processing with reduced power consumption.
[0170] In the ca...
Embodiment 3
[0196] In Embodiment 3, a computer 100 that realizes the reserve pool calculation of Embodiment 1 using an optical circuit chip will be described.
[0197] Figure 8 It is a figure showing an example of the structure of the optical circuit chip of Example 3. in addition, Figure 8 Corresponding to the top view of the optical circuit chip.
[0198] The optical circuit chip 800 mounts a plurality of functional chips on a substrate 801 . In addition, since the optical circuit is mounted in the stacking direction relative to the electronic circuit, optical elements such as the MZ modulator and photodiode do not appear in the drawings.
[0199] The optical circuit chip 800 includes a substrate 801, a silicon nitride optical circuit 802, a silicon optical circuit 803, a substrate 804, a sampling circuit 805, a masking circuit 806, a delay circuit 807, a modulator driving circuit 808, a recursive signal amplifier 809, and a transimpedance amplifier 810 , a readout circuit 811 and...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More