Efficient Rare Event Sampling with Unsupervised Normalising Flows

Abstract

From Physics and Biology to Seismology and Economics, the behaviour of countless systems is determined by impactful yet unlikely transitions between metastable states known as \emph{rare events}, the study of which is essential for understanding and controlling these systems' properties. Classical computational methods to sample rare events remain prohibitively inefficient, and are a bottleneck for enhanced samplers requiring prior data. Here, we introduce a novel framework, FlowRES that uses unsupervised normalising flow neural networks to enhance Monte Carlo sampling of rare events by generating high-quality nonlocal Monte Carlo proposals. We validate FlowRES by sampling the transition path ensembles of equilibrium and non-equilibrium systems of Brownian particles exploring increasingly complex potential surfaces. Beyond eliminating requirements for prior data, FlowRES features key advantages over established samplers: no collective variables need defining, its efficiency remains constant even as events become increasingly unlikely, and it can handle systems with multiple routes between states.

References in corpus (18)

• Escaping free-energy minima
• Active Particles in Complex and Crowded Environments
• An efficient, multiple range random walk algorithm to calculate the density of states
• Normalizing Flows: An Introduction and Review of Current Methods
• Active Brownian Particles. From Individual to Collective Stochastic Dynamics
• Sampling rare switching events in biochemical networks
• Unsupervised speech representation learning using WaveNet autoencoders
• The topography of the environment alters the optimal search strategy for active particles
• Machine learning strategies for path-planning microswimmers in turbulent flows
• Studying Rare Events using Forward-Flux Sampling: Recent Breakthroughs and Future Outlook
• Perturbative calculation of quasi-potential in non-equilibrium diffusions: a mean-field example
• Metadynamics of paths
• Practical guide to replica exchange transition interface sampling and forward flux sampling
• Target search of active agents crossing high energy barriers
• Efficient path sampling on multiple reaction channels
• Learning how to find targets in the micro-world: The case of intermittent active Brownian particles
• Optimal foraging strategies can be learned
• Scalable Normalizing Flows Enable Boltzmann Generators for Macromolecules

Cited by in corpus (4)

• Policy-guided Monte Carlo on general state spaces: Application to glass-forming mixtures
• Towards neural reinforcement learning for large deviations in nonequilibrium systems with memory
• When velocity autocorrelations mirror force autocorrelations: Exact noise-cancellation in interacting Brownian systems
• Sampling the Liquid-Gas Critical Point with Boltzmann Generators