# US 2020/0401651 A1 · Information processing device and optimization method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2020/0401651 A1 - Document ID: US20200401651A1 - Status: Published application - Kind: A1 - Application number: 16/886,890 - Filed: 2020-05-29 - Published: 2020-12-24 - Earliest claimed: 2019-06-18 - Assignees: Fujitsu Limited - Applicants: Fujitsu Limited - Inventors: Hirotaka TAMURA; Makiko Konoshima - Web page: https://patentgrep.com/patent/US20200401651A1 ## CPC - G06F 17/11: for solving equations {, e.g. nonlinear equations, general mathematical optimization problems (optimization specially adapted for a specific administrative, business or logistic context G06Q10/04)} - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/047: Probabilistic or stochastic networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An information processing device includes: a memory configured to hold values of state variables included in an evaluation function presenting energy and a weight value for each set of the state variables; and a processor coupled to the memory and configured to: calculate an energy change value when each of the values of the state variables is set as a next change candidate based on the values of the state variables and the weight value; calculate a total energy change value by adding a penalty value according to an excess amount violating an inequality constraint, to each of the energy change values calculated for the state variables, the excess amount being calculated based on a coupling coefficient and a threshold value; and change any value of the state variables in the memory based on a set temperature value, a random number value, and the total energy change values. ## Claims ### Claim 1 An information processing device comprising: a memory configured to hold values of a plurality of state variables included in an evaluation function representing energy and a weight value for each set of the state variables; and a processor coupled to the memory and configured to: calculate an energy change value when each of the values of the plurality of state variables is set as a next change candidate based on the values of the plurality of state variables and the weight value, in a case where any value of the plurality of state variables changes; calculate a total energy change value by adding a penalty value according to an excess amount violating an inequality constraint, to each of a plurality of the energy change values calculated for the plurality of state variables, the excess amount being calculated based on a coupling coefficient indicating a weight of each of the plurality of state variables in the inequality constraint and a threshold value; and change any value of the plurality of state variables held in the memory based on a set temperature value, a random number value, and a plurality of the total energy change values. ### Claim 2 (depends on 1) The information processing device according to claim 1, wherein the processor calculates the penalty value by multiplying the change value of the excess amount calculated based on the coupling coefficient and the threshold value by an importance coefficient indicating importance of the inequality constraint. ### Claim 3 (depends on 1) The information processing device according to claim 1, wherein, when the inequality constraint is an inequality constraint for an upper limit constraint value, the processor adds a first penalty value according to a change value of a first excess amount, which is calculated based on a first coupling coefficient and a first threshold value, in a positive direction for the first threshold value, to each of the plurality of energy change values. ### Claim 4 (depends on 1) The information processing device according to claim 1, wherein, when the inequality constraint is an inequality constraint for a lower limit constraint value, the processor adds a second penalty value according to a change value of a second excess amount, which is calculated based on a second coupling coefficient and a second threshold value, in a negative direction for the second threshold value, to each of the plurality of energy change values. ### Claim 5 (depends on 1) The information processing device according to claim 1, wherein, when the inequality constraint is an inequality constraint for an upper limit constraint value and a lower limit constraint value, the processor adds a third penalty value according to a larger change value between a change value of a third excess amount, which is calculated based on a third coupling coefficient and a third threshold value, in a positive direction for the third threshold value, and a change value of a fourth excess amount, which is calculated based on the third coupling coefficient and a fourth threshold value smaller than the third threshold value, in a negative direction for the fourth threshold value, to each of the plurality of energy change values. ### Claim 6 (depends on 1) The information processing device according to claim 1, wherein, in a case of an equality constraint for a predetermined constraint value, the processor adds a fourth penalty value according to a larger change value between a change value of a fifth excess amount, which is calculated based on a fourth coupling coefficient and a fifth threshold value, in a positive direction for the fifth threshold value, and a change value of a sixth excess amount in a negative direction for the fifth threshold value, to each of the plurality of energy change values. ### Claim 7 (depends on 1) The information processing device according to claim 1, wherein the memory holds the coupling coefficient for a set of a state variable and an inequality constraint variable corresponding to the inequality constraint, and wherein the coupling coefficient is an asymmetric coupling coefficient in which a weight of the state variable in a local field of the inequality constraint variable is set to a value other than 0 and a weight of the inequality constraint variable in a local field of the state variable is set to 0. ### Claim 8 (depends on 1) The information processing device according to claim 1, wherein the processor selects any change value of the plurality of energy change values, based on identification information of a state variable of a change target, and calculates an energy value corresponding to values of the plurality of state variables held in the memory by accumulating the selected change value. ### Claim 9 (depends on 1) The information processing device according to claim 1, wherein the processor is provided as plural and the temperature values different from each other are set to the corresponding processor and the temperature values or the values of the plurality of state variables after reaching a number of repetitions of a ground state search or after a certain time elapses are exchanged. ### Claim 10 An optimization method comprising: calculating, by a computer, an energy change value when each of values of a plurality of state variables is set as a next change candidate based on the values of the plurality of state variables and a weight value held in a memory, in a case where any value of the plurality of state variables held in the memory changes; calculating a total energy change value by adding a penalty value according to an excess amount violating an inequality constraint to each of a plurality of the energy change values calculated for the plurality of state variables, the excess amount being calculated based on a coupling coefficient indicating a weight of each of the plurality of state variables in the inequality constraint and a threshold value; and changing any value of the plurality of state variables held in the memory based on a set temperature value, a random number value, and a plurality of the total energy change values.