Mixed Integer Neural Inverse Design
This addresses the challenge of inverse design for computational designers and engineers, offering a principled method for combinatorial tasks, though it is incremental as it builds on existing neural network surrogates.
The paper tackles the inverse design problem in computational design by formulating it as a mixed-integer linear program, leveraging the piecewise linear property of neural networks to find globally optimal or near-optimal solutions, with applications in material selection and robustness to fabrication perturbations.
In computational design and fabrication, neural networks are becoming important surrogates for bulky forward simulations. A long-standing, intertwined question is that of inverse design: how to compute a design that satisfies a desired target performance? Here, we show that the piecewise linear property, very common in everyday neural networks, allows for an inverse design formulation based on mixed-integer linear programming. Our mixed-integer inverse design uncovers globally optimal or near optimal solutions in a principled manner. Furthermore, our method significantly facilitates emerging, but challenging, combinatorial inverse design tasks, such as material selection. For problems where finding the optimal solution is not desirable or tractable, we develop an efficient yet near-optimal hybrid optimization. Eventually, our method is able to find solutions provably robust to possible fabrication perturbations among multiple designs with similar performances.