Arithmetic processing device and system to realize multi-layer convolutional neural network circuit to perform process with fixed-point number format

An arithmetic processing system to realize a multi-layer convolutional neural network circuit to perform in a process with fixed-point number format, comprising a first device, a second device, and a memory

• the first device conducting:
• a learning process to perform weight learning or bias learning using learning data stored in the memory to calculate initial weight values and initial bias values for layers of the multi-layer convolutional neural network circuit;
• a trial recognition process to perform a recognition process to part of the learning data or of input data using the initial weight values and the initial bias values;
• and a processing treatment process to multiply the initial weight values and the initial bias values by a positive constant depending on the layers to calculate processed weight values and processed bias values for the layers, wherein the positive constant is determined from an intermediate value in a process of each layer, the intermediate value being obtained as a result of the recognition process of the trial recognition process, and
• the second device conducting a recognition process using the processed weight values and the processed bias values.