Abstract: Inverse molecular design aims to discover novel tailored materials, given a desired functionality. Recent advances from the rapidly growing field of artificial intelligence, mostly from the subfield of machine learning, have resulted in a fertile exchange of ideas, where new approaches to inverse molecular design are being proposed and employed at a rapid pace.

Among these, deep generative models show promise in modeling the space of possible chemicals in a probabilistic way, allowing us to interpolate and optimize molecules. We will look at two approaches for making generative models, ​the variational auto encoder and adversarial training with reinforcement learning, along with examples of their application.