MACHINE LEARNING SYSTEM, MACHINE LEARNING DEVICE AND MACHINE LEARNING METHOD
원문보기
IPC분류정보
국가/구분 |
United States(US) Patent
공개
|
국제특허분류(IPC7판) |
|
출원번호 |
17915353
(2021-01-15)
|
공개번호 |
20230144616
(2023-05-11)
|
우선권정보 |
JP-2020-078883 (2020-04-28); JP-2020-116497 (2020-07-06) |
국제출원번호 |
PCT/JP2021/001234
(2021-01-15)
|
발명자
/ 주소 |
- TAKI, Yuko
- TAKATSUKA, Susumu
- TETSUKAWA, Hiroki
|
출원인 / 주소 |
|
인용정보 |
피인용 횟수 :
0 인용 특허 :
0 |
초록
▼
There is provided a machine learning system, a machine learning device and a machine learning method, which respectively prompts a person to engage in a target behavior with machine learning on a correlation between such a person's behavior and an environment around them. The machine learning system
There is provided a machine learning system, a machine learning device and a machine learning method, which respectively prompts a person to engage in a target behavior with machine learning on a correlation between such a person's behavior and an environment around them. The machine learning system includes at least: a state acquisition unit that acquires at least state information regarding a behavior of a person; an evaluation unit that obtains a value function by evaluating environment information regarding an environment around the person at the time of acquiring the state information and the state information; and a machine learning classifier that performs reinforcement learning on the value function in order to prompt the person to engage in a target behavior and selects the environment information when the value function becomes highest.
대표청구항
▼
1. A machine learning system, comprising at least: a state acquisition unit configured to acquire at least state information regarding a behavior of a person;an evaluation unit configured to obtain a value function by evaluating the state information and environment information regarding an environm
1. A machine learning system, comprising at least: a state acquisition unit configured to acquire at least state information regarding a behavior of a person;an evaluation unit configured to obtain a value function by evaluating the state information and environment information regarding an environment around the person when acquiring the state information; anda machine learning classifier that performs reinforcement learning on the value function and selects the environment information when the value function is the highest in order to prompt the person to engage in a target behavior. 2. The machine learning system according to claim 1, wherein the evaluation unit is configuredto calculate a reward on a basis of a difference between the state information and target state information regarding the target behavior, andto calculate the value function on a basis of the reward, the environment information and the state information. 3. The machine learning system according to claim 1, wherein the system holds target state-related information including a plurality of pieces of target behavior information. 4. The machine learning system according to claim 3, wherein the target state-related information includes time-specific target state information and/or stage-specific target state information. 5. The machine learning system according to claim 1, wherein the environment information includes information regarding scents, lighting, temperature, humidity, video or sound. 6. The machine learning system according to claim 1, further comprising a scent control unit,wherein the scent control unit is configured to control generated scent on a basis of the environment information selected by the machine learning classifier. 7. The machine learning system according to claim 6, further comprising an aromatization unit,wherein the aromatization unit is configured to make items have scent on a basis of the environment information selected by the machine learning classifier, andthe machine learning classifier determines which of the scent control unit and the aromatization unit will generate scent on a basis of the environment information. 8. The machine learning system according to claim 1, further comprising a lighting control unit,wherein the lighting control unit is configured to control light to be emitted on a basis of the environment information selected by the machine learning classifier. 9. The machine learning system according to claim 1, further comprising an air conditioning unit,wherein the air conditioning unit is configured to control a temperature and/or humidity on a basis of the environment information selected by the machine learning classifier. 10. The machine learning system according to claim 1, further comprising a video control unit,wherein the video control unit is configured to control a video to be displayed on a basis of the environment information selected by the machine learning classifier. 11. The machine learning system according to claim 1, further comprising a sound control unit,wherein the sound control unit is configured to control a sound to be played on a basis of the environment information selected by the machine learning classifier. 12. The machine learning system according to claim 1, wherein the value function is divided into a plurality of value groups, andthe machine learning classifier uses the value function held by each of the plurality of value groups. 13. The machine learning system according to claim 1, further comprising: a plurality of state acquisition units; andan achievement difficulty level calculation unit,wherein the achievement difficulty levelcalculation unit is configured to calculate anachievement difficulty level for the target behavior on a basis of the state information acquired by each of the plurality of state acquisition units. 14. The machine learning system according to claim 13, wherein the achievement difficulty level includes an achievement rate indicating a degree to which the target behavior is prompted. 15. The machine learning system according to claim 13, wherein the achievement difficulty level includes a standard achievement time indicating a standard time for which the target behavior is prompted. 16. The machine learning system according to claim 13, wherein the achievement difficulty level includes a number of key variables indicating an average number of items in the environment information when the target behavior is prompted. 17. A machine learning device, comprising at least: a state acquisition unit configured to acquire at least state information regarding a behavior of a person;an evaluation unit configured to obtain a value function by evaluating the state information and environment information regarding an environment around the person when acquiring the state information; anda machine learning classifier that performs reinforcement learning on the value function and selects the environment information when the value function is the highest in order to prompt the person to engage in a target behavior. 18. A machine learning method, comprising at least: acquiring at least state information regarding a behavior of a person;obtaining a value function by evaluating the state information and environment information regarding an environment around the person when acquiring the state information; andperforming reinforcement learning on the value function and selecting the environment information when the value function is the highest in order to prompt the person to engage in a target behavior.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.