Doru C. Lupascu

Boris N. Slautin, Utkarsh Pratiush, Doru C. Lupascu et al.

We introduce a Deep Kernel Learning Variational Autoencoder (VAE-DKL) framework that integrates the generative power of a Variational Autoencoder (VAE) with the predictive nature of Deep Kernel Learning (DKL). The VAE learns a latent representation of high-dimensional data, enabling the generation of novel structures, while DKL refines this latent space by structuring it in alignment with target properties through Gaussian Process (GP) regression. This approach preserves the generative capabilities of the VAE while enhancing its latent space for GP-based property prediction. We evaluate the framework on two datasets: a structured card dataset with predefined variational factors and the QM9 molecular dataset, where enthalpy serves as the target function for optimization. The model demonstrates high-precision property prediction and enables the generation of novel out-of-training subset structures with desired characteristics. The VAE-DKL framework offers a promising approach for high-throughput material discovery and molecular design, balancing structured latent space organization with generative flexibility.

Boris N. Slautin, Kamyar Barakati, Hiroshi Funakubo et al.

Microscopy techniques generate vast amounts of complex image data that in principle can be used to discover simpler, interpretable, and parsimonious forms to reveal the underlying physical structures, such as elementary building blocks in molecular systems or order parameters and phases in crystalline materials. Variational Autoencoders (VAEs) provide a powerful means of constructing such low-dimensional representations, but their performance heavily depends on multiple non-myopic design choices, which are often optimized through trial-and-error and empirical analysis. To enable automated and unbiased optimization of VAE workflows, we investigated reward-based strategies for evaluating latent space representations. Using Piezoresponse Force Microscopy data as a model system, we examined multiple policies and reward functions that can serve as a foundation for automated optimization. Our analysis shows that approximating the latent space with Gaussian Mixture Models (GMM) and Bayesian Gaussian Mixture Models (BGMM) provides a strong basis for constructing reward functions capable of estimating model efficiency and guiding the search for optimal parsimonious representations.