Title:
TASK ALLOCATION DEVICE, METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2023/243047
Kind Code:
A1
Abstract:
A task allocation device according to one embodiment comprises: an allocation unit which inputs, to a neural network, information indicative of a feature of each of a plurality of agents that are subjects of operations to a plurality of tasks and information indicative of a feature of each of the plurality of tasks, and generates, on the basis of this input result, information indicative of the allocation of the tasks to the agents; a generation unit which generates a response to an operation, which is performed by the agent with respect to the task, on the basis of the information generated by the allocation unit; and a training unit which evaluates whether the solution to an operation of the agent to the task is good, the solution being generated on the basis of information that is training information for parameters of the neural network and is generated by the allocation unit with respect to the information indicative of the feature of each of the plurality of agents and the information indicative of the feature of each of the plurality of tasks, and performs a training process of updating the parameters of the neural network in response to the result of this evaluation.
Inventors:
AKASHI KAZUAKI (JP)
KANAI SHUNSUKE (JP)
SATO MASATAKA (JP)
OGAWA MANAMI (JP)
YAMAZOE MAYU (JP)
KANAI SHUNSUKE (JP)
SATO MASATAKA (JP)
OGAWA MANAMI (JP)
YAMAZOE MAYU (JP)
Application Number:
PCT/JP2022/024163
Publication Date:
December 21, 2023
Filing Date:
June 16, 2022
Export Citation:
Assignee:
NIPPON TELEGRAPH & TELEPHONE (JP)
International Classes:
G06Q10/04
Domestic Patent References:
| WO2021106977A1 | 2021-06-03 | |||
| WO2022038773A1 | 2022-02-24 |
Other References:
AKASHI, KAZUAKI: "A Study of Reinforcement Learning for Solving Multi-Vehicle Routing Problems", IEICE TECHNICAL REPORT, vol. 121, no. 13 (ICM2021-5), 6 May 2021 (2021-05-06), pages 23 - 28, XP009551609
MATSUBAYASHI, HIROAKI: "B-14-8 Calculation of rescue building priority order in consideration of network topology", PROCEEDINGS OF THE 2021 IEICE COMMUNICATIONS SOCIETY CONFERENCE; SEPTEMBER 14-17, 2021, vol. 2021, no. 2, 31 August 2021 (2021-08-31) - 17 September 2021 (2021-09-17), pages 182, XP009551611
KAZUHIRO AKASHI, SHUNSUKE KANAI, SATOSHI SUZUKI, KENICHI TAYAMA: "B-14-10 Evaluation of delivery proposal methods that use reinforced learning in emergencies", PROCEEDINGS OF THE 2020 IEICE SOCIETY CONFERENCE, 2; 2020.09.15-18, IEICE, JP, 1 September 2020 (2020-09-01) - 18 September 2020 (2020-09-18), JP, pages 168, XP009551234
MATSUBAYASHI, HIROAKI: "B-14-8 Calculation of rescue building priority order in consideration of network topology", PROCEEDINGS OF THE 2021 IEICE COMMUNICATIONS SOCIETY CONFERENCE; SEPTEMBER 14-17, 2021, vol. 2021, no. 2, 31 August 2021 (2021-08-31) - 17 September 2021 (2021-09-17), pages 182, XP009551611
KAZUHIRO AKASHI, SHUNSUKE KANAI, SATOSHI SUZUKI, KENICHI TAYAMA: "B-14-10 Evaluation of delivery proposal methods that use reinforced learning in emergencies", PROCEEDINGS OF THE 2020 IEICE SOCIETY CONFERENCE, 2; 2020.09.15-18, IEICE, JP, 1 September 2020 (2020-09-01) - 18 September 2020 (2020-09-18), JP, pages 168, XP009551234
Attorney, Agent or Firm:
SUZUYE & SUZUYE (JP)
Download PDF: