Developments and applications of orbital-based learning as a general and accurate property predictor
Project Description
Orbital-based learning methods are a class of methods using atomic orbital (AO) or molecular orbital (MO) descriptors computed from lower level theory as inputs to learn higher quality molecular properties. This class of methods has emerged as a crucial approach for developing highly accurate electronic structure theories. In our previous work, we have introduced two distinct orbital-based learning approaches (MOB-ML & OrbNet), each tailored for specific scales and utilizing different representations computed from varying computational levels. In this project, we would like to (1) try to unify and improve the two orbital-based learning approaches using Deep Kernel Learning (DKL) as a whole idea to bridge the equivarient graph neural network (EGNN) or graph Transformer as a general representation learner and Gaussian Process Regression (GPR) as a regressor, to fully use date, (2) apply current orbital-based learning as a general potential predictor to real applications in various chemical problems, (3) construct a general pretrained chemical representation via orbital representations using Contrastive learning
Supervisor
CHENG, Lixue
Quota
2
Course type
UROP1000
UROP1100
UROP2100
UROP3100
UROP4100
Applicant's Roles
(1) Compute molecular property data from high quality wavefunction theory
(2) Train deep learning models using orbital-based learning methods
(3) Develop and improve the current model architecture
(4) Able to apply the models to real-world problems, such as ML potential and molecular predictions.
(2) Train deep learning models using orbital-based learning methods
(3) Develop and improve the current model architecture
(4) Able to apply the models to real-world problems, such as ML potential and molecular predictions.
Applicant's Learning Objectives
(1) Be familiar with our current codebase and methodology of OrbNet (AO+EGNN) and MOB-ML (MO+GPR)
(2) Be familiar with basic electronic structure theories, including Hatree-Fock, Coupled-cluster, GFN-xTB, and commonly used DFTs
(3) Be familiar with PySCF and able to run PySCF for feature and label generations
(4) Be able to train deep learning models on GPUs
(5) Could assist the new model architecture improvements
(6) Could understand how to bridge computational chemistry and quantum simulations with real chemistry systems and applications
(2) Be familiar with basic electronic structure theories, including Hatree-Fock, Coupled-cluster, GFN-xTB, and commonly used DFTs
(3) Be familiar with PySCF and able to run PySCF for feature and label generations
(4) Be able to train deep learning models on GPUs
(5) Could assist the new model architecture improvements
(6) Could understand how to bridge computational chemistry and quantum simulations with real chemistry systems and applications
Complexity of the project
Challenging