A CQM-based approach to solving a combinatorial problem with applications in drug design

Abstract

The use of D-Wave's Leap Hybrid solver is demonstrated here in solving a Knapsack optimization problem: finding meal combinations from a fixed menu that fit a diner's constraints. This is done by first formulating the optimization problem as a Constrained Quadratic Model (CQM) and then submitting it to a quantum annealer. We highlight here the steps needed, as well as the implemented code, and provide solutions from a Chicken and Waffle restaurant menu. Additionally, we discuss how this model may be generalized to find optimal drug molecules within a large search space with many complex, and often contradictory, structures and property constraints.