Multifidelity multilevel Monte Carlo to accelerate approximate Bayesian parameter inference for partially observed stochastic processes
arXiv:2110.14082 · doi:10.1016/j.jcp.2022.111543
Abstract
Models of stochastic processes are widely used in almost all fields of science. Theory validation, parameter estimation, and prediction all require model calibration and statistical inference using data. However, data are almost always incomplete observations of reality. This leads to a great challenge for statistical inference because the likelihood function will be intractable for almost all partially observed stochastic processes. This renders many statistical methods, especially within a Bayesian framework, impossible to implement. Therefore, computationally expensive likelihood-free approaches are applied that replace likelihood evaluations with realisations of the model and observation process. For accurate inference, however, likelihood-free techniques may require millions of expensive stochastic simulations. To address this challenge, we develop a new method based on recent advances in multilevel and multifidelity. Our approach combines the multilevel Monte Carlo telescoping summation, applied to a sequence of approximate Bayesian posterior targets, with a multifidelity rejection sampler to minimise the number of computationally expensive exact simulations required for accurate inference. We present the derivation of our new algorithm for likelihood-free Bayesian inference, discuss practical implementation details, and demonstrate substantial performance improvements. Using examples from systems biology, we demonstrate improvements of more than two orders of magnitude over standard rejection sampling techniques. Our approach is generally applicable to accelerate other sampling schemes, such as sequential Monte Carlo, to enable feasible Bayesian analysis for realistic practical applications in physics, chemistry, biology, epidemiology, ecology and economics.
References in corpus (30)
- Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
- The frontier of simulation-based inference
- The pseudo-marginal approach for efficient Monte Carlo computations
- Adaptive approximate Bayesian computation
- Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification
- A modified Next Reaction Method for simulating chemical systems with time dependent propensities and delays
- Approximation and inference methods for stochastic biochemical kinetics - a tutorial review
- The statistical mechanics of complex signaling networks : nerve growth factor signaling
- On the utility of graphics cards to perform massively parallel simulation of advanced Monte Carlo methods
- Multilevel Sequential Monte Carlo Samplers
- Simulation and inference algorithms for stochastic biochemical reaction networks: from basic concepts to state-of-the-art
- Flexible statistical inference for mechanistic models of neural dynamics
- Approximate Bayesian Computation for Forward Modeling in Cosmology
- Transport map accelerated Markov chain Monte Carlo
- Error analysis of tau-leap simulation methods
- Variational Bayes with Synthetic Likelihood
- Delayed acceptance particle MCMC for exact inference in stochastic kinetic models
- Multilevel Sequential Monte Carlo for Bayesian Inverse Problems
- An adaptive multi-level simulation algorithm for stochastic biological systems
- Multilevel rejection sampling for approximate Bayesian computation
- Extending the multi-level method for the simulation of stochastic biological systems
- SIMD Parallel MCMC Sampling with Applications for Big-Data Bayesian Analytics
- A practical guide to pseudo-marginal methods for computational inference in systems biology
- Multilevel Markov Chain Monte Carlo Method for High-Contrast Single-Phase Flow Problems
- Approximate Bayesian Model Inversion for PDEs with Heterogeneous and State-Dependent Coefficients
- Rapid Bayesian inference for expensive stochastic models
- Multifidelity Approximate Bayesian Computation
- Quasi-Monte Carlo methods applied to tau-leaping in stochastic biological systems
- Bayesian estimation of incompletely observed diffusions
- Vector operations for accelerating expensive Bayesian computations -- a tutorial guide