# US 2021/0240439 A1 · Arithmetic processing device, arithmetic processing method, and non-transitory computer-readable storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0240439 A1 - Document ID: US20210240439A1 - Status: Published application - Kind: A1 - Application number: 17/130,122 - Filed: 2020-12-22 - Published: 2021-08-05 - Earliest claimed: 2020-02-04 - Assignees: Fujitsu Limited - Applicants: Fujitsu Limited - Inventors: Yi Ge; Katsuhiro Yoda; Makiko Ito - Web page: https://patentgrep.com/patent/US20210240439A1 ## CPC - G06F 7/38: Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation - G06F 9/30014: {with variable precision} - G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/0495: Quantised networks; Sparse networks; Compressed networks - G06N 3/063: using electronic means - G06N 3/08: Learning methods - G06N 3/084: Backpropagation, e.g. using gradient descent - G06N 3/09: Supervised learning - G06N 3/045: Combinations of networks - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An arithmetic processing device includes a memory and a processor coupled to the memory. The processor configured to calculate statistical information of a first operation result by executing the predetermined operation using input data as a first fixed-point number with a first decimal point at a first decimal point position, determine a second decimal point position using the statistical information, and calculate a second operation result when the predetermined operation is executed using the input data as a second fixed-point number with a second decimal point at the second decimal point position. ## Claims ### Claim 1 An arithmetic processing device comprising: a memory; and a processor coupled to the memory and configured to: calculate statistical information of a first operation result by executing a predetermined operation using input data as a first fixed-point number with a first decimal point at a first decimal point position, determine a second decimal point position using the statistical information, and specify a second operation result when the predetermined operation is executed using the input data as a second fixed-point number with a second decimal point at the second decimal point position. ### Claim 2 (depends on 1) The arithmetic processing device according to claim 1, wherein the processor executes a first process acquiring the second operation result by executing the predetermined operation using the input data as the second fixed-point number. ### Claim 3 (depends on 2) The arithmetic processing device according to claim 2, wherein the processor includes a first sub processor and a second sub processor, the first sub processor executes the predetermined operation using the input data as the first fixed-point number, and the second sub processor executes the predetermined operation using the input data as the second fixed-point number. ### Claim 4 (depends on 1) The arithmetic processing device according to claim 1, wherein the memory stores the first operation result, and processor executes a second process acquiring the second operation result by converting the first operation result stored in the memory into the second fixed-point number with the second decimal point at the second decimal point position. ### Claim 5 (depends on 4) The arithmetic processing device according to claim 4, wherein the processor includes a first sub processor and a second sub processor, the second sub processor executes the predetermined operation using the input data as the first fixed-point number with the first decimal point at the first decimal point position, and the first sub processor executes a process of converting the first operation result stored in the memory into the second fixed-point number with the second decimal point at the second decimal point position. ### Claim 6 (depends on 1) The arithmetic processing device according to claim 1, wherein the memory configured to store the first operation result, the processor selects a process to be executed for a shorter time period from a first process and a second process, the first process acquiring the second operation result by causing the operation unit to execute the predetermined operation using the input data as the second fixed-point number with the second decimal point at the second decimal point position, the second process acquiring the second operation result by converting the first operation result stored in the memory into the second fixed-point number with the second decimal point at the second decimal point position and executes the selected process. ### Claim 7 (depends on 6) The arithmetic processing device according to claim 6, wherein the processor selects, based on the type of the predetermined operation, the process to be executed for a shorter time period from the first process and the second process. ### Claim 8 (depends on 2) The arithmetic processing device according to claim 2, wherein the processor executes the predetermined operation using the first operation data being input as the first fixed-point number with the first decimal point at the first decimal point position, and acquires the statistical information, the first operation data being a part of a plurality of operation data items included in the input data. ### Claim 9 (depends on 4) The arithmetic processing device according to claim 4 wherein the executes the predetermined operation using a first operation data being input as the first fixed-point number with the first decimal point at the first decimal point position and calculate the first operation result, and acquires the statistical information, the first operation data being a part of a plurality of operation data items included in the input data, and when the second process is executed, the processor converts the first operation result into the second fixed-point number with the second decimal point at the second decimal point position, inputs second operation data included in the input data and excluding the first operation data, execute the predetermined operation using the second operation data as the second fixed-point number with the second decimal point at the second decimal point position, and acquires the second operation result. ### Claim 10 (depends on 8) The arithmetic processing device according to claim 8, wherein the processor changes a ratio of the first operation data to the input data. ### Claim 11 (depends on 1) The arithmetic processing device according to claim 1, wherein the input data is an image data. ### Claim 12 An arithmetic processing method comprising: calculating statistical information of a first operation result by causing to execute a predetermined operation using input data as a first fixed-point number with a first decimal point at a first decimal point position; determining a second decimal point position using the statistical information; and calculating a second operation result when the predetermined operation is executed using the input data as a second fixed-point number with a second decimal point at the second decimal point position. ### Claim 13 A non-transitory computer-readable storage medium storing a program that causes a processor included in a noise estimation apparatus to execute a process, the process comprising: calculating statistical information of a first operation result by causing to execute a predetermined operation using input data as a first fixed-point number with a first decimal point at a first decimal point position; determining a second decimal point position using the statistical information; and calculating a second operation result when the predetermined operation is executed using the input data as a second fixed-point number with a second decimal point at the second decimal point position.