| CPC G06F 30/27 (2020.01) [G06N 3/08 (2013.01); G06N 20/00 (2019.01); H01F 41/04 (2013.01); G06F 30/373 (2020.01)] | 20 Claims |

|
1. A method for generating a candidate inductor design comprising:
receiving target specifications for an inductor;
generating an inductor design segment-by-segment using a reinforcement learning agent to generate segment parameters for each added segment, wherein the reinforcement learning agent implements a policy that is learned using a reward computed based on performance of the generated inductor design relative to the target specifications, wherein the inductor design is generated segment-by-segment by:
inputting to the reinforcement learning agent a current inductor state representing any existing segments of the inductor design;
obtaining, using the policy implemented by the reinforcement learning agent and based on the inputted current inductor state, segment parameters for a new segment to be added to the inductor design;
updating the current inductor state to connect the new segment to any existing segments in the inductor design, in accordance with the segment parameters obtained from the reinforcement learning agent; and
repeating the inputting, obtaining and updating until a termination condition is met; and
outputting the generated inductor design as the candidate inductor design after determining that the generated inductor design satisfies a predefined performance threshold.
|