Method and apparatus for reinforcement machine learning

A method for performing exclusive reinforcement learning by an agent device, the method comprising:

• collecting information of states of an environment and performing a statistical analysis on the states using the collected information;
• determining a first state value of a first state among the states in a training phase and a second state value of a second state among the states in an inference phase based on analysis results of the statistical analysis;
• performing reinforcement learning by using a plurality of reinforcement learning unit which performs reinforcement learnings from different perspectives according to the first state value;
• and
• selecting one of actions determined by the plurality of reinforcement learning unit based on the second state value and applying selected action to the environment,
• wherein the plurality of reinforcement learning unit includes a central perspective reinforcement learning unit and a peripheral perspective reinforcement learning unit, and
• wherein performing reinforcement learning by using a plurality of reinforcement learning unit which performs reinforcement learnings from different perspectives according to the first state value comprises
• performing the reinforcement learning by using the peripheral perspective reinforcement learning unit or the central perspective reinforcement learning unit according to the first state value,
• wherein the agent device applies the selected action to the environment and the agent device receives a reward from the environment, where the reward is input to the central perspective reinforcement learning unit and the peripheral perspective reinforcement learning unit, respectively, and
• wherein:
• the analysis result of the statist analysis includes an average and a standard deviation of collected sample states,
• determining a first slate value of a state among tin states in a training phase and a second state value of a second state among the states in an inference phase based on analysis results of the statistical analysis comprises
• determining the second state value to 1 when an absolute value of the second state is greater than a sum of the average and the standard deviation and determining the second state value to 0 when the absolute value of the second state is than or equal to the sum of the average and the standard deviation,
• the plurality of reinforcement learning unit includes a central perspective reinforcement learning unit and a peripheral perspective reinforcement learning unit, and
• selecting one of actions determined plurality of reinforcement earning, unit based on the second and applying selected action to the environment comprises
• selecting an action determined by the peripheral perspective reinforcement learning unit is when the second state value is 1 and selecting an action determined by the central perspective reinforcement learning unit then the second stat value is 0.

An agent apparatus for performing exclusive reinforcement learning, the apparatus comprising:

• processor, memory, and communication interface,
• wherein the processor executes the program stored in the memory to perform:
• collecting information of states of an environment through the communication interface and performing a statistical analysis on the states using the collected information;
• determining a first state value of a first state among the states in a training phase and a second state value of a second state among the states in an inference phase based on analysis results of the statistical analysis;
• performing reinforcement learning by using one reinforcement learning unit of a plurality of reinforcement learning unit which performs reinforcement learnings from different perspectives according to the first state value;
• and
• selecting one of actions determined by the plurality of reinforcement learning unit based on the second state value and applying selected action to the environment,
• wherein the plurality of reinforcement learning unit includes a central perspective reinforcement learning unit and a peripheral perspective reinforcement learning unit, and
• when the processor performs the step of performing reinforcement learning by using a plurality of reinforcement learning unit which performs reinforcement learnings from different perspectives according to the first state value, the processor performs
• performing the reinforcement learning by using the peripheral perspective reinforcement learning unit or the central perspective reinforcement learning unit according to the first state value,
• wherein the agent device applies the selected action to the environment and the agent device receives a reward from the environment, where the reward is input to the central perspective reinforcement learning unit and the peripheral perspective reinforcement learning unit, respectively, and
• wherein:
• the analysis result of the statistical analysis includes an average and a standard deviation of collected sample states,
• when the processor performs the step of determining a first state value of a first state among the states in a training phase and a second state value of a second state among the states in an inference phase based on analysis results of the statistical analysis, the processor performs
• determining the second state value to 1 when an absolute value of the second state is greater than a sum of the average and the standard deviation and determining the second state value to 0 when the absolute value of the second state is less than r equal to the sum of the average and the standard deviation,
• the plurality of reinforcement learning unit includes a central perspective reinforcement learning unit and a peripheral perspective reinforcement learning unit, and
• when the processor performs the step of selecting one of actions determined by the plurality of reinforcement learning unit based on the second state value and applying selected action to the environment, the processor performs
• selecting an action determined by the peripheral perspective reinforcement learning unit when the second state value is 1 and selecting an action determined by the central perspective reinforcement learning unit when the second state value is 0.