Challenges in natural sciences can often be phrased as optimization problems. Ma-chine learning techniques have recently been applied to solve such problems. One example in chemistry is the design of tailor-made organic materials and molecules, which requires efficient methods to explore the chemical space. We present a genetic algorithm (GA) that is enhanced with a neural network (DNN) based dis-criminator model to improve the diversity of generated molecules and at the same time steer the GA. We show that our algorithm outperforms other generative mod-els in optimization tasks. We furthermore present a way to increase interpretability of genetic algorithms, which helped us to derive design principles.
View on arXiv