Method of generating device model and computing device performing the same

A method of generating a device model configured to perform a simulation based on device data and output simulation result data indicating characteristics of a semiconductor device, the method being performed by at least one processor executing program codes, the method comprising:

• receiving measurement data obtained by measuring characteristics of the semiconductor device;
• selecting at least one target parameters among a plurality of parameters of the device model configured to perform the simulation;
• selecting a plurality of initial value sets corresponding to different combinations of initial values of the at least one target parameter;
• determining a plurality of local minimum values based on reinforcement learning such that each of the plurality of local minimum values corresponds to a minimum value of a difference between the measurement data and the simulation result data with respect to each of the plurality of initial value sets;
• performing a genetic algorithm such that a plurality of global minimum values equal to or smaller than the plurality of local minimum values is determined;
• and
• determining improved values of the at least one target parameter based on the plurality of global minimum values.

A computing device comprising:

• a computer readable medium storing program codes including a training control module configured to train a device model, the device model configured to perform a simulation based on device data and output simulation result data indicating characteristics of a semiconductor device;
• and
• at least one processor configured to, by executing the training control module,
• receive measurement data obtained by measuring the characteristics of the semiconductor device;
• select at least one target parameter among a plurality of parameters of the device model;
• select a plurality of initial value sets corresponding to different combinations of initial values of the at least one target parameter;
• determine a plurality of local minimum values based on reinforcement learning such that each of the plurality of local minimum value corresponds to a minimum value of a difference between the measurement data and the simulation result data with respect to each of the plurality of initial value sets;
• and
• determine improved values of the at least one target parameters based on the plurality of local minimum values.