Topological feature engineering for machine learning based halide perovskite materials design

Abstract

Accelerated materials development with machine learning (ML) assisted screening and high throughput experimentation for new photovoltaic materials holds the key to addressing our grand energy challenges. Data-driven ML is envisaged as a decisive enabler for new perovskite materials discovery. However, its full potential can be severely curtailed by poorly represented molecular descriptors (or fingerprints). Optimal descriptors are essential for establishing effective mathematical representations of quantitative structure-property relationships. Here we reveal that our persistent functions (PFs) based learning models offer significant accuracy advantages over traditional descriptor based models in organic-inorganic halide perovskite (OIHP) materials design and have similar performance as deep learning models. Our multiscale simplicial complex approach not only provides a more precise representation for OIHP structures and underlying interactions, but also has better transferability to ML models. Our results demonstrate that advanced geometrical and topological invariants are highly efficient feature engineering approaches that can markedly improve the performance of learning models for molecular data analysis. Further, new structure-property relationships can be established between our invariants and bandgaps. We anticipate that our molecular representations and featurization models will transcend the limitations of conventional approaches and lead to breakthroughs in perovskite materials design and discovery.

Machine Learning for Perovskite Solar Cells and Component Materials: Key Technologies and Prospects

“…The selection of which features to use as inputs for the models can greatly affect the performance and interpretability of ML models. [ 254 ] Although there are methods that can help to select the number of inputs (see Section 2.2.1), these methods also have limitations, such as not taking into account the potential physicochemical links between features and not being able to select the types of features. This may weaken the interpretability and credibility of the model.…”

Carriers, Quasi-particles, and Collective Excitations in Halide Perovskites

“…Underpinning HPs’ intriguing optoelectronic properties are their unusual electronic structures, which have been extensively investigated using DFT. ,, Of late, they have also been combined with machine learning approaches to design and tune their fascinating properties . While DFT calculations severely underestimate the bandgap, an approach using quasi-particle self-consistent GW with spin–orbit coupling corrections has been found to provide consistent and reliable results for HPs. ,, Figure a shows the electronic band structure of the archetypical 3D perovskite CH 3 NH 3 PbI 3 (MAPbI 3 ).…”

Phase Transitions and Dynamics in Mixed Three- and Low-Dimensional Lead Halide Perovskites

“…The temperature effects and dynamics can be probed using MD, − but such simulations may also suffer from the pronounced finite-size effects, especially if ab initio MD is used. The MC simulations are capable of simulating large many-particle systems and temperature effects, but usually the atomistic picture must be simplified to a coarse-grained model based on the effective model Hamiltonian formalism. ,,− Recently, significant attention was also concentrated on the machine-learning-augmented calculations, which are capable of predicting new structures, phase transitions, and properties of (mixed) lead halide perovskites, while overcoming the need for large-scale simulations. ,,− …”