Abstract: Over the last 10 years, functional electronic materials design has undergone a shift from chemical-intuition-based strategies to data-driven synthesis and simulation. Numerous machine learning models have been developed to successfully predict materials properties and generate new crystal structures. Many existing approaches, however, rely upon physical insights to construct handcrafted features (descriptors), which are not always readily available. For novel materials with sparse prior data and insufficient physical understanding, conventional machine learning models may display limited predictability.

In this talk, I will address this challenge by introducing an adaptive optimization engine for materials composition optimization, where feature engineering is not explicitly required—so called featureless learning. I then describe a use case where multi-objective Bayesian optimization with latent-variable Gaussian processes is utilized to accelerate the design of electronic metal-insulator transition compounds for memory applications. I will then contrast this approach with supervised classification-based models for MIT compounds. Last, I will highlight a recent quantitative study on structure-property relationship in crystal systems enabled by deep neural networks. The model, which learns the structural genome, identifies intrinsically similar structures in Fourier space, and reveals that some crystal structures inherently host high propensities for optimal materials properties. This finding enables the decoupling of structure and composition for future codesign of multifunctionality. Finally, I propose how integration of these different modalities could lead to harmonious iterative exploration of novel functional materials.