Optimization learns from the user

A method to incorporate a user preference into a multi-objective optimization while the multi-objective optimization is running, the method comprising:

• performing the multi-objective optimization by,
• generating a first plurality of intermediate designs based on optimizing a plurality of variable values corresponding to the multiple dimensions of the multi-objective optimization, wherein each of the first plurality of intermediate designs includes a corresponding weight,
• outputting the first plurality of it designs, wherein the first plurality of intermediate designs are presented in a user interface to a user concurrently with the multi-objective optimization generating additional intermediate designs,
• receiving a preference indicator for a selected one of the first plurality of intermediate designs, the preference indicator inputted by the user through the user interface concurrently with the multi-objective optimization generating additional intermediate designs, wherein at least some of the additional intermediate designs are a least two design generations ahead of the selected one of the first plurality of intermediate design,
• adjusting the corresponding weight of the selected one of the first plurality of intermediate designs based on the preference indicator, and
• generating a second plurality of intermediate designs using at least the some of the additional designs and the adjusted corresponding weight of the selected one of the first plurality of intermediate designs, wherein a design with a greater weight will have a greater number of mutations generated in the second plurality of intermediate designs than a design with a lower weight.

A device that incorporates a user preference into a multi-objective optimization while the multi-objective optimization is running, the device comprising:

• a processor;
• a memory coupled to the processor through a bus;
• and
• a process executed from the memory by the processor that causes the processor to perform the multi-objective optimization, wherein the process further causes the processor to generate a first plurality of intermediate designs based on optimizing a plurality of variable values corresponding to the multiple dimensions of the multi-objective optimization, wherein each of the first plurality of intermediate designs includes a corresponding weight, output the first plurality of intermediate designs, wherein the first plurality of intermediate designs are presented in a user interface to a user concurrently with the multi-objective optimization generating additional intermediate designs, receive a preference indicator for a selected one of the first plurality of intermediate designs, the preference indicator inputted by the user through the user interface concurrently with the multi-objective optimization generating additional intermediate designs, wherein at least some of the additional it designs are at least two design generations ahead of the selected one of the first plurality of intermediate design, adjust the corresponding weight of the selected one of the first plurality of intermediate designs based on the preference indicator, and generate a second plurality of intermediate designs using at least the some of the additional designs and the adjusted corresponding weight of the selected one of the first plurality of intermediate designs, wherein a design with a greater weight will have a greater number of mutations generated in the second plurality of intermediate designs than a design with a lower weight.