Predictive collective variable discovery with deep Bayesian models
Abstract
Extending spatio-temporal scale limitations of models for complex atomistic systems considered in biochemistry and materials science necessitates the development of enhanced sampling methods. The potential acceleration in exploring the configurational space by enhanced sampling methods depends on the choice of collective variables (CVs). In this work, we formulate the discovery of CVs as a Bayesian inference problem and consider the CVs as hidden generators of the full-atomistic trajectory. The ability to generate samples of the fine-scale atomistic configurations using limited training data allows us to compute estimates of observables as well as our probabilistic confidence on them. The methodology is based on emerging methodological advances in machine learning and variational inference. The discovered CVs are related to physicochemical properties which are essential for understanding mechanisms especially in unexplored complex systems. We provide a quantitative assessment of the CVs in terms of their predictive ability for alanine dipeptide (ALA-2) and ALA-15 peptide.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Jan 14, 2019
- Source ID
- 10.1063/1.5058063
Entities
People
- Markus Schöberl
- Nicholas Zabaras
- Phaedon-Stelios Koutsourelakis
Organizations
- Defense Advanced Research Projects Agency
- Technical University of Munich
- University of Notre Dame