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.

References in corpus (18)

• Adam: A Method for Stochastic Optimization
• Auto-Encoding Variational Bayes
• Escaping free-energy minima
• Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method
• Tackling Exascale Software Challenges in Molecular Dynamics Simulations with GROMACS
• Searching for Activation Functions
• Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
• A Variational Approach to Enhanced Sampling and Free Energy Calculations
• Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration
• Improving Variational Inference with Inverse Autoregressive Flow
• A variational conformational dynamics approach to the selection of collective variables in metadynamics
• Identification of simple reaction coordinates from complex dynamics
• Accelerating Minibatch Stochastic Gradient Descent using Stratified Sampling
• Path-space variational inference for non-equilibrium coarse-grained systems
• Black Box Variational Inference
• Predictive Coarse-Graining
• Multilevel coarse graining and nano--pattern discovery in many particle stochastic systems
• Physics-constrained, data-driven discovery of coarse-grained dynamics

Cited by in corpus (15)

• Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks
• Data-Driven Collective Variables for Enhanced Sampling
• Neural networks-based variationally enhanced sampling
• A unified framework for machine learning collective variables for enhanced sampling simulations: $\text{mlcolvar}$
• Teaching the Incompressible Navier-Stokes Equations to Fast Neural Surrogate Models in 3D
• Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications
• Ensuring thermodynamic consistency with invertible coarse-graining
• Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
• A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables
• Chemistrees: data driven identification of reaction pathways via machine learning
• Learning Incompressible Fluid Dynamics from Scratch -- Towards Fast, Differentiable Fluid Models that Generalize
• Physics-Constrained Predictive Molecular Latent Space Discovery with Graph Scattering Variational Autoencoder
• Bayesian stochastic multi-scale analysis via energy considerations
• Machine learning approaches for analyzing and enhancing molecular dynamics simulations
• Variational embedding of protein folding simulations using gaussian mixture variational autoencoders