arXiv subjects

Dennis Prangle

Publications and source records attributed to Dennis Prangle.

At least 19 recordsLinked to original sources

arXiv2026

Minimum Distance Summaries for Robust Neural Posterior Estimation

Simulation-based inference (SBI) enables amortized Bayesian inference by first training a neural posterior estimator (NPE) on prior-simulator pairs, typically through low-dimensional summary statistics, which can then be cheaply reused for fast inference by querying it on new test observations. Because NPE is estimated under the training data distribution, it is susceptible to misspecification when observations deviate from the training distribution. Many robust SBI approaches address this by modifying NPE training or introducing error models, coupling robustness to the inference network and compromising amortization and modularity. We introduce minimum-distance summaries, a plug-in robust NPE method that adapts queried test-time summaries independently of the pretrained NPE. Leveraging the maximum mean discrepancy (MMD) as a distance between observed data and a summary-conditional predictive distribution, the adapted summary inherits strong robustness properties from the MMD. We demonstrate that the algorithm can be implemented efficiently with random Fourier feature approximations, yielding a lightweight, model-free test-time adaptation procedure. We provide theoretical guarantees for the robustness of our algorithm and empirically evaluate it on a range of synthetic and real-world tasks, demonstrating substantial robustness gains with minimal additional overhead.

stat.ML↗

arXiv2025

A Comparison of Kernels for ABC-SMC

A popular method for likelihood-free inference is approximate Bayesian computation sequential Monte Carlo (ABC-SMC) algorithms. These approximate the posterior using a population of particles, which are updated using Markov kernels. Several such kernels have been proposed. In this paper we review these, highlighting some less well known choices, and proposing some novel options. Further, we conduct an extensive empirical comparison of kernel choices. Our results suggest using a one-hit kernel with a mixture proposal as a default choice.

stat.CO↗

arXiv2024

Bayesian Emulation of Geotechnical Deterioration Curves Using Quadratic and B-Spline Hierarchical Models

The stability of geotechnical infrastructure assets, such as cuttings and embankments, is crucial to the safe and efficient delivery of transport services. The successful emulation of geotechnical models of deterioration of infrastructure slopes has the potential to inform slope design, maintenance and remediation by introducing the time dependency of deterioration into geotechnical asset management. We have performed computer experiments of deterioration, measured by the factor of safety (FoS), for a set of cutting slope geometries and soil properties that are common in the southern UK. Whilst computer experiments are an extremely useful and cost-effective method of better understanding deterioration mechanisms, it would not be practical to run enough experiments to understand relations between high-dimensional inputs and outputs. Therefore, we trained a fully-Bayesian Gaussian process emulator using an ensemble of 75 computer experiments to predict the FoS. We construct two different emulator models, one approximating the FoS temporal evolution with a quadratic model and one approximating the temporal evolution with a B-spline model; and we emulated their parameters. We also compare the ability of our models to predict failure time. The developed models could be used to inform infrastructure cutting slope design and management, and extend serviceable life.

stat.AP↗

arXiv2024

Flexible Tails for Normalizing Flows

Normalizing flows are a flexible class of probability distributions, expressed as transformations of a simple base distribution. A limitation of standard normalizing flows is representing distributions with heavy tails, which arise in applications to both density estimation and variational inference. A popular current solution to this problem is to use a heavy tailed base distribution. We argue this can lead to poor performance due to the difficulty of optimising neural networks, such as normalizing flows, under heavy tailed input. We propose an alternative, "tail transform flow" (TTF), which uses a Gaussian base distribution and a final transformation layer which can produce heavy tails. Experimental results show this approach outperforms current methods, especially when the target distribution has large dimension or tail weight.

stat.ML↗

arXiv2024

Optimal combination of composite likelihoods using approximate Bayesian computation with application to state-space models

Composite likelihood provides approximate inference when the full likelihood is intractable and sub-likelihood functions of marginal events can be evaluated relatively easily. It has been successfully applied for many complex models. However, its wider application is limited by two issues. First, weight selection of marginal likelihood can have a significant impact on the information efficiency and is currently an open question. Second, calibrated Bayesian inference with composite likelihood requires curvature adjustment which is difficult for dependent data. This work shows that approximate Bayesian computation (ABC) can properly address these two issues by using multiple composite score functions as summary statistics. First, the summary-based posterior distribution gives the optimal Godambe information among a wide class of estimators defined by linear combinations of estimating functions. Second, to make ABC computationally feasible for models where marginal likelihoods have no closed form, a novel approach is proposed to estimate all simulated marginal scores using a Monte Carlo sample with size N. Sufficient conditions are given for the additional noise to be negligible with N fixed as the data size n goes to infinity, and the computational cost is O(n). Third, asymptotic properties of ABC with summary statistics having heterogeneous convergence rates is derived, and an adaptive scheme to choose the component composite scores is proposed. Numerical studies show that the new method significantly outperforms the existing Bayesian composite likelihood methods, and the efficiency of adaptively combined composite scores well approximates the efficiency of particle MCMC using the full likelihood.

stat.ME↗

arXiv2023

Flexible Tails for Normalising Flows, with Application to the Modelling of Financial Return Data

We propose a transformation capable of altering the tail properties of a distribution, motivated by extreme value theory, which can be used as a layer in a normalizing flow to approximate multivariate heavy tailed distributions. We apply this approach to model financial returns, capturing potentially extreme shocks that arise in such data. The trained models can be used directly to generate new synthetic sets of potentially extreme returns

stat.ML↗

arXiv2020

Measure Transport with Kernel Stein Discrepancy

Measure transport underpins several recent algorithms for posterior approximation in the Bayesian context, wherein a transport map is sought to minimise the Kullback--Leibler divergence (KLD) from the posterior to the approximation. The KLD is a strong mode of convergence, requiring absolute continuity of measures and placing restrictions on which transport maps can be permitted. Here we propose to minimise a kernel Stein discrepancy (KSD) instead, requiring only that the set of transport maps is dense in an $L^2$ sense and demonstrating how this condition can be validated. The consistency of the associated posterior approximation is established and empirical results suggest that KSD is competitive and more flexible alternative to KLD for measure transport.

stat.CO↗

arXiv2020

Bonsai-Net: One-Shot Neural Architecture Search via Differentiable Pruners

One-shot Neural Architecture Search (NAS) aims to minimize the computational expense of discovering state-of-the-art models. However, in the past year attention has been drawn to the comparable performance of naive random search across the same search spaces used by leading NAS algorithms. To address this, we explore the effects of drastically relaxing the NAS search space, and we present Bonsai-Net, an efficient one-shot NAS method to explore our relaxed search space. Bonsai-Net is built around a modified differential pruner and can consistently discover state-of-the-art architectures that are significantly better than random search with fewer parameters than other state-of-the-art methods. Additionally, Bonsai-Net performs simultaneous model search and training, dramatically reducing the total time it takes to generate fully-trained models from scratch.

cs.LG↗

arXiv2019

Distilling Importance Sampling for Likelihood Free Inference

Likelihood-free inference involves inferring parameter values given observed data and a simulator model. The simulator is computer code which takes parameters, performs stochastic calculations, and outputs simulated data. In this work, we view the simulator as a function whose inputs are (1) the parameters and (2) a vector of pseudo-random draws. We attempt to infer all these inputs conditional on the observations. This is challenging as the resulting posterior can be high dimensional and involve strong dependence. We approximate the posterior using normalizing flows, a flexible parametric family of densities. Training data is generated by likelihood-free importance sampling with a large bandwidth value epsilon, which makes the target similar to the prior. The training data is "distilled" by using it to train an updated normalizing flow. The process is iterated, using the updated flow as the importance sampling proposal, and slowly reducing epsilon so the target becomes closer to the posterior. Unlike most other likelihood-free methods, we avoid the need to reduce data to low dimensional summary statistics, and hence can achieve more accurate results. We illustrate our method in two challenging examples, on queuing and epidemiology.

stat.CO↗

arXiv2019

The Neural Moving Average Model for Scalable Variational Inference of State Space Models

Variational inference has had great success in scaling approximate Bayesian inference to big data by exploiting mini-batch training. To date, however, this strategy has been most applicable to models of independent data. We propose an extension to state space models of time series data based on a novel generative model for latent temporal states: the neural moving average model. This permits a subsequence to be sampled without drawing from the entire distribution, enabling training iterations to use mini-batches of the time series at low computational cost. We illustrate our method on autoregressive, Lotka-Volterra, FitzHugh-Nagumo and stochastic volatility models, achieving accurate parameter estimation in a short time.

stat.ML↗

arXiv2019

Black-Box Inference for Non-Linear Latent Force Models

Latent force models are systems whereby there is a mechanistic model describing the dynamics of the system state, with some unknown forcing term that is approximated with a Gaussian process. If such dynamics are non-linear, it can be difficult to estimate the posterior state and forcing term jointly, particularly when there are system parameters that also need estimating. This paper uses black-box variational inference to jointly estimate the posterior, designing a multivariate extension to local inverse autoregressive flows as a flexible approximater of the system. We compare estimates on systems where the posterior is known, demonstrating the effectiveness of the approximation, and apply to problems with non-linear dynamics, multi-output systems and models with non-Gaussian likelihoods.

stat.ML↗

arXiv2019

Ensemble MCMC: Accelerating Pseudo-Marginal MCMC for State Space Models using the Ensemble Kalman Filter

Particle Markov chain Monte Carlo (pMCMC) is now a popular method for performing Bayesian statistical inference on challenging state space models (SSMs) with unknown static parameters. It uses a particle filter (PF) at each iteration of an MCMC algorithm to unbiasedly estimate the likelihood for a given static parameter value. However, pMCMC can be computationally intensive when a large number of particles in the PF is required, such as when the data is highly informative, the model is misspecified and/or the time series is long. In this paper we exploit the ensemble Kalman filter (EnKF) developed in the data assimilation literature to speed up pMCMC. We replace the unbiased PF likelihood with the biased EnKF likelihood estimate within MCMC to sample over the space of the static parameter. On a wide class of different non-linear SSM models, we demonstrate that our new ensemble MCMC (eMCMC) method can significantly reduce the computational cost whilst maintaining reasonable accuracy. We also propose several extensions of the vanilla eMCMC algorithm to further improve computational efficiency. Computer code to implement our methods on all the examples can be downloaded from https://github.com/cdrovandi/Ensemble-MCMC.

stat.CO↗

arXiv2019

Bayesian experimental design without posterior calculations: an adversarial approach

Most computational approaches to Bayesian experimental design require making posterior calculations repeatedly for a large number of potential designs and/or simulated datasets. This can be expensive and prohibit scaling up these methods to models with many parameters, or designs with many unknowns to select. We introduce an efficient alternative approach without posterior calculations, based on optimising the expected trace of the Fisher information, as discussed by Walker (2016). We illustrate drawbacks of this approach, including lack of invariance to reparameterisation and encouraging designs in which one parameter combination is inferred accurately but not any others. We show these can be avoided by using an adversarial approach: the experimenter must select their design while a critic attempts to select the least favourable parameterisation. We present theoretical properties of this approach and show it can be used with gradient based optimisation methods to find designs efficiently in practice.

stat.CO↗

arXiv2019

Optimality Criteria for Probabilistic Numerical Methods

It is well understood that Bayesian decision theory and average case analysis are essentially identical. However, if one is interested in performing uncertainty quantification for a numerical task, it can be argued that standard approaches from the decision-theoretic framework are neither appropriate nor sufficient. Instead, we consider a particular optimality criterion from Bayesian experimental design and study its implied optimal information in the numerical context. This information is demonstrated to differ, in general, from the information that would be used in an average-case-optimal numerical method. The explicit connection to Bayesian experimental design suggests several distinct regimes in which optimal probabilistic numerical methods can be developed.

stat.ME↗

arXiv2018

Black-Box Autoregressive Density Estimation for State-Space Models

State-space models (SSMs) provide a flexible framework for modelling time-series data. Consequently, SSMs are ubiquitously applied in areas such as engineering, econometrics and epidemiology. In this paper we provide a fast approach for approximate Bayesian inference in SSMs using the tools of deep learning and variational inference.

stat.ML↗

arXiv2018

Black-box Variational Inference for Stochastic Differential Equations

Parameter inference for stochastic differential equations is challenging due to the presence of a latent diffusion process. Working with an Euler-Maruyama discretisation for the diffusion, we use variational inference to jointly learn the parameters and the diffusion paths. We use a standard mean-field variational approximation of the parameter posterior, and introduce a recurrent neural network to approximate the posterior for the diffusion paths conditional on the parameters. This neural network learns how to provide Gaussian state transitions which bridge between observations in a very similar way to the conditioned diffusion process. The resulting black-box inference method can be applied to any SDE system with light tuning requirements. We illustrate the method on a Lotka-Volterra system and an epidemic model, producing accurate parameter estimates in a few hours.

stat.CO↗

arXiv2017

Marginal sequential Monte Carlo for doubly intractable models

Bayesian inference for models that have an intractable partition function is known as a doubly intractable problem, where standard Monte Carlo methods are not applicable. The past decade has seen the development of auxiliary variable Monte Carlo techniques (M{\o}ller et al., 2006; Murray et al., 2006) for tackling this problem; these approaches being members of the more general class of pseudo-marginal, or exact-approximate, Monte Carlo algorithms (Andrieu and Roberts, 2009), which make use of unbiased estimates of intractable posteriors. Everitt et al. (2017) investigated the use of exact-approximate importance sampling (IS) and sequential Monte Carlo (SMC) in doubly intractable problems, but focussed only on SMC algorithms that used data-point tempering. This paper describes SMC samplers that may use alternative sequences of distributions, and describes ways in which likelihood estimates may be improved adaptively as the algorithm progresses, building on ideas from Moores et al. (2015). This approach is compared with a number of alternative algorithms for doubly intractable problems, including approximate Bayesian computation (ABC), which we show is closely related to the method of M{\o}ller et al. (2006).

stat.CO↗

arXiv2017

gk: An R Package for the g-and-k and generalised g-and-h Distributions

The g-and-k and (generalised) g-and-h distributions are flexible univariate distributions which can model highly skewed or heavy tailed data through only four parameters: location and scale, and two shape parameters influencing the skewness and kurtosis. These distributions have the unusual property that they are defined through their quantile function (inverse cumulative distribution function) and their density is unavailable in closed form, which makes parameter inference complicated. This paper presents the gk R package to work with these distributions. It provides the usual distribution functions and several algorithms for inference of independent identically distributed data, including the finite difference stochastic approximation method, which has not been used before for this problem.

stat.CO↗