A machine learning apparatus includes: a state observation unit that observes a state variable including an error between a position command and an actual position of a rotor, temperature of a motor driving apparatus and the motor, and voltage of each part of the motor driving apparatus; and a learning unit that learns a current feedback offset correction value for correcting an offset in the current feedback value, an inter-current-feedback-phase unbalance correction value for correcting an unbalance between phases in the current feedback value, and a current command correction value for a dead zone for correcting a current command in order to compensate a decreased amount of current due to a dead zone by which switching elements of upper and lower arms in the same phase of an inverter for motor power supply are not simultaneously turned on, in accordance with a training data set defined by the state variable.