Data-informed Deep Optimization

TL;DR

This paper introduces a data-informed deep optimization method that uses neural networks to learn feasible regions and optimize high-dimensional design problems efficiently with limited data.

Contribution

The paper presents a novel DiDo approach combining DNN classifiers and surrogate modeling for high-dimensional, data-scarce optimization problems.

Findings

Effective in practical industrial design case with limited data

Successfully applied to a 100-dimensional toy problem

Demonstrates flexibility and promise for high-dimensional optimization

Abstract

Complex design problems are common in the scientific and industrial fields. In practice, objective functions or constraints of these problems often do not have explicit formulas, and can be estimated only at a set of sampling points through experiments or simulations. Such optimization problems are especially challenging when design parameters are high-dimensional due to the curse of dimensionality. In this work, we propose a data-informed deep optimization (DiDo) approach as follows: first, we use a deep neural network (DNN) classifier to learn the feasible region; second, we sample feasible points based on the DNN classifier for fitting of the objective function; finally, we find optimal points of the DNN-surrogate optimization problem by gradient descent. To demonstrate the effectiveness of our DiDo approach, we consider a practical design case in industry, in which our approach yields good solutions using limited size of training data. We further use a 100-dimension toy example to show the effectiveness of our model for higher dimensional problems. Our results indicate that the DiDo approach empowered by DNN is flexible and promising for solving general high-dimensional design problems in practice.