Active Importance Sampling for Variational Objectives Dominated by Rare Events: Consequences for Optimization and Generalization

Abstract

Deep neural networks, when optimized with sufficient data, provide accurate representations of high-dimensional functions; in contrast, function approximation techniques that have predominated in scientific computing do not scale well with dimensionality. As a result, many high-dimensional sampling and approximation problems once thought intractable are being revisited through the lens of machine learning. While the promise of unparalleled accuracy may suggest a renaissance for applications that require parameterizing representations of complex systems, in many applications gathering sufficient data to develop such a representation remains a significant challenge. Here we introduce an approach that combines rare events sampling techniques with neural network optimization to optimize objective functions that are dominated by rare events. We show that importance sampling reduces the asymptotic variance of the solution to a learning problem, suggesting benefits for generalization. We study our algorithm in the context of learning dynamical transition pathways between two states of a system, a problem with applications in statistical physics and implications in machine learning theory. Our numerical experiments demonstrate that we can successfully learn even with the compounding difficulties of high-dimension and rare data.

References in corpus (8)

• Machine learning and the physical sciences
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• A Stochastic Gradient Method with an Exponential Convergence Rate for Finite Training Sets
• Solving Many-Electron Schrödinger Equation Using Deep Neural Networks
• Sampling Can Be Faster Than Optimization
• Exact dynamical coarse-graining without time-scale separation
• Fluctuation-dissipation relations for stochastic gradient descent
• Solving for high dimensional committor functions using artificial neural networks

Cited by in corpus (7)

• Reinforcement learning of rare diffusive dynamics
• Inexact iterative numerical linear algebra for neural network-based spectral estimation and rare-event prediction
• Variational deep learning of equilibrium transition path ensembles
• Supervised Learning and the Finite-Temperature String Method for Computing Committor Functions and Reaction Rates
• Consensus between Epistemic Agents is Difficult
• From high-dimensional committors to reactive insights
• Committor functions via tensor networks