Evolutionary Approach to Constructing a Deep Feedforward Neural Network for Prediction of Electronic Coupling Elements in Molecular Materials

Abstract

We present a general framework for the construction of a deep feedforward neural network (FFNN) to predict distance and orientation dependent electronic coupling elements in disordered molecular materials. An evolutionary algorithm automatizes the selection of an optimal architecture of the artificial neural network within a predefined search space. Systematic guidance, beyond minimizing the model error with stochastic gradient descent based backpropagation, is provided by simultaneous maximization of a model fitness that takes into account additional physical properties, such as the field-dependent carrier mobility. As a prototypical system, we consider hole transport in amorphous tris(8-hydroxyquinolinato)aluminum. Reference data for training and validation is obtained from multiscale ab initio simulations, in which coupling elements are evaluated using density-functional theory, for a system containing 4096 molecules. The Coulomb matrix representation is chosen to encode the explicit molecular pair coordinates into a rotation and translation invariant feature set for the FFNN. The final optimized deep feedforward neural network is tested for transport models without and with energetic disorder. It predicts electronic coupling elements and mobilities in excellent agreement with the reference data. Such a FFNN is readily applicable to much larger systems at negligible computational cost, providing a powerful surrogate model to overcome the size limitations of the ab initio approach.

Machine-Learned Charge Transfer Integrals for Multiscale Simulations in Organic Thin Films

“…Within the context of charge transport simulations, ML has been successfully used to predict reorganization energies and transfer integrals − in different organic semiconductors. In our previous work, we used kernel ridge regression (KRR) to predict transfer integrals in thermally disordered pentacene crystals using system specific geometric features .…”

Machine Learning of Quasiparticle Energies in Molecules and Clusters

“…The application of Kernel Ridge Regression (KRR) models to predicting molecular quantum properties has been very successful over recent years [8,14,15,20,41]. The main idea relies on constructing a kernel matrix with a kernel function k that can quantitatively measure similarity between molecular representations x i and x j , which are vector representations that encode the molecular physics [6,21,24]. The Laplacian kernel function, for example, is described as…”

“…Hence, the incorporation of quantum machine learning (QML) has been gaining great traction over recent years. QML based surrogate property models have become a popular alternative approach for their fast, reliable, and accurate predictions of molecular and material properties [1][2][3][4][5][6][7][8][9][10][11][12][13][14].…”

“…The main advantage of ML models is that they allow predictions of molecular properties with improved efficiency at a lower computational cost compared to traditional quantum chemistry approaches. Method development in the field of QML is progressing rapidly and it is increasingly influencing traditional methods [6,[15][16][17][18]. Developments in molecular representations and QML models have paved the way for predicting energetic, electronic, and thermodynamic properties, such as atomization energies, dipole moments, polarizabilities, and harmonic frequencies [19][20][21].…”

Data-Driven Modeling of S0 -> S1 Excitation Energy in the BODIPY Chemical Space: High-Throughput Computation, Quantum Machine Learning, and Inverse Design

“…For excited state properties, in general, the error rates in QML have been noted to be inferior compared to that of ground state properties [61][62][63][64]. Yet, QML methods continue to find applications in excited state modeling in chemical space datasets [50,[65][66][67] as well as in potential surface manifolds [68][69][70][71][72][73]. Keeping abreast with the progress in QML, materials/molecules inverse-design protocols have also advanced since the earliest implementation nearly twenty years ago [74].…”