Reinforcement learning to allocate processes to a machine tool controller

A machine learning device that performs reinforcement learning on a controller that performs multiple processes in parallel at multiple operation units for controlling a machine tool, the controller including a first multi-core control unit exerting asymmetrical multi-processing (AMP) control of fixing an operation unit to perform a process and a second multi-core control unit exerting symmetrical multi-processing (SMP) control of not fixing an operation unit to perform a process, the machine learning device comprising:

• a memory configured to store a program;
• and
• a processor configured to execute the program and control the machine learning device to:
• output behavior information containing allocation of arithmetic units that perform the multiple processes to the controller;
• acquire state information containing a machining condition as a condition for machining set at the machine tool and determination information generated by monitoring an implementation of the multiple processes by the multiple operation units based on the allocation in the behavior information;
• calculate a value of a reward to be given by the reinforcement learning based on the determination information in the state information;
• and
• update a behavior value function based on the value of a reward, the state information, and the behavior information.