LiteGEM: Lite Geometry Enhanced Molecular Representation Learning for Quantum Property Prediction

TL;DR

LiteGEM is a novel graph neural network-based approach that leverages geometry-enhanced molecular representations and self-supervised learning to accurately predict quantum properties like the HOMO-LUMO gap.

Contribution

This work introduces LiteGEM, a new geometry-enhanced molecular representation learning framework that improves quantum property prediction accuracy on large-scale datasets.

Findings

Achieved MAE of 0.1204 on test set

Utilized deep graph neural networks with self-supervised tasks

Effective in large-scale quantum chemistry prediction

Abstract

In this report, we (SuperHelix team) present our solution to KDD Cup 2021-PCQM4M-LSC, a large-scale quantum chemistry dataset on predicting HOMO-LUMO gap of molecules. Our solution, Lite Geometry Enhanced Molecular representation learning (LiteGEM) achieves a mean absolute error (MAE) of 0.1204 on the test set with the help of deep graph neural networks and various self-supervised learning tasks. The code of the framework can be found in https://github.com/PaddlePaddle/PaddleHelix/tree/dev/competition/kddcup2021-PCQM4M-LSC/.