2012
NeurIPS
NeurIPS 2012
Learning Invariant Representations of Molecules for Atomization Energy Prediction
Abstract
The accurate prediction of molecular energetics in chemical compound space is a crucial ingredient for rational compound design. The inherently graph-like, non-vectorial nature of molecular data gives rise to a unique and difficult machine learning problem. In this paper, we adopt a learning-from-scratch approach where quantum-mechanical molecular energies are predicted directly from the raw molecular geometry. The study suggests a benefit from setting flexible priors and enforcing invariance stochastically rather than structurally. Our results improve the state-of-the-art by a factor of almost three, bringing statistical methods one step closer to the holy grail of ''chemical accuracy''.
🌉
Interdisciplinary Bridge
- Artificial Intelligence and Healthcare & Medicine and Interdisciplinary
📈
Trend Setter
- Feature Learning
🧭
Keyword Pioneer
- graph neural network
🐣
Hot Topic Early Bird
- graph neural network
Authors
Topics
Machine Learning > Core Methods > Representation Learning
Machine Learning > Learning Types > Self-Supervised Learning
Deep Learning > Architectures > Graph Neural Networks
Healthcare & Medicine > Research > Bioinformatics
Machine Learning > Bayesian & Probabilistic > Bayesian Learning
Deep Learning > Learning Types > Representation Learning
Interdisciplinary > Science > Bioinformatics
Deep Learning > Learning Types > Feature Learning
Keywords
graph representation
molecular machine learning
quantum mechanical energies
graph-like molecular data
atomization energy prediction
molecular geometry
atomization energy
graph-like data
molecular representation
invariant representation
molecular property prediction
molecular representation learning
graph neural network
quantum-mechanical energy
chemical accuracy