Parameter Inference Research Articles

Tree sequences as a general-purpose tool for population genetic inference.

As population genetics data increases in size new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient, but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks (CNNs) applied to population genetic alignments. To better utilize these new data structures we propose and implement a graph convolutional network (GCN) to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard CNN approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a GCN approach and can be used to perform well on these common population genetics inference tasks with accuracies roughly matching or even exceeding that of a CNN-based method. As tree sequences become more widely used in population genetics research we foresee developments and optimizations of this work to provide a foundation for population genetics inference moving forward.

Hierarchical Bayesian data selection

There are many issues that can cause problems when attempting to infer model parameters from data. Data and models are both imperfect, and as such there are multiple scenarios in which standard methods of inference will lead to misleading conclusions; corrupted data, models which are only representative of subsets of the data, or multiple regions in which the model is best fit using different parameters. Methods exist for the exclusion of some anomalous types of data, but in practice, data cleaning is often undertaken by hand before attempting to fit models to data. In this work, we will employ hierarchical Bayesian data selection; the simultaneous inference of both model parameters, and parameters which represent our belief that each observation within the data should be included in the inference. The aim, within a Bayesian setting, is to find the regions of observation space for which the model can well-represent the data, and to find the corresponding model parameters for those regions. A number of approaches will be explored, and applied to test problems in linear regression, and to the problem of fitting an ODE model, approximated by a finite difference method. The approaches are simple to implement, can aid mixing of Markov chains, and are broadly applicable to many inferential problems.

Addressing challenges associated with training sample sizes for effective calibration of NWP precipitation forecasts

Abstract Effective calibration of precipitation forecasts produced by numerical weather prediction (NWP) models faces challenges associated with the training sample size. Newly-operationalized NWP models may only accumulate a small number of forecasts and thus may limit robust parameter inference in forecast calibration. It is necessary to investigate how the performance of forecast calibration changes with the amount of training data, to determine an effective training sample size. In this study, we thoroughly investigate the impacts of training sample size on precipitation forecast calibration based on the Seasonally Coherent Calibration (SCC) model across Australia. Overall, the performance of the model tends to stabilize in most parts of Australia when raw forecasts of 10 months or longer are used for parameter inference. Whether the training dataset cover wet months substantially affects forecast calibration. The findings of this study are critical for understanding the impacts of training sample size on forecast calibration, and will provide implications for future forecast calibration and the generation of hindcasts.

Tree sequences as a general-purpose tool for population genetic inference.

As population genetics data increases in size new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient, but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks (CNNs) applied to population genetic alignments. To better utilize these new data structures we propose and implement a graph convolutional network (GCN) to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard CNN approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a GCN approach and can be used to perform well on these common population genetics inference tasks with accuracies roughly matching or even exceeding that of a CNN-based method. As tree sequences become more widely used in population genetics research we foresee developments and optimizations of this work to provide a foundation for population genetics inference moving forward.

Inference of the acoustic properties of transversely isotropic porous materials

Transversely isotropic porous materials are well represented by equivalent fluid models based on semi-phenomenological models such as the Johnson-Champoux-Allard-Lafarge (JCAL) model and a set of material parameters. Direct measurements of these parameters, however, are often difficult or time consuming. As an alternative, inverse estimation methods provide a simple and fast framework to identify the JCAL model parameters, since they can be performed on single impedance tube measurements. In this work, we present a stochastic inverse identification method of the JCAL parameters for transversely isotropic materials. We consider the flow resistance and characteristic viscous length and the static tortuosity to be anisotropic and the frame to be fully rigid, resulting in nine unique parameters to be identified. The method is based on reflection factor measurements of a material in different orientations corresponding the transverse and isotropic principal axes. Hence, a priori knowledge of the orientation of the principal axes of the material is required. Results of the parameter inference are presented based on experimental data for a glass wool sample.

Bridging Hydrological Ensemble Simulation and Learning Using Deep Neural Operators

AbstractEnsemble‐based simulation and learning (ESnL) has long been used in hydrology for parameter inference, but computational demands of process‐based ESnL can be quite high. To address this issue, we propose a deep neural operator learning approach. Neural operators are generic machine learning algorithms that can learn functional mappings between infinite‐dimensional spaces, providing a highly flexible tool for scientific machine learning. Our approach is built upon DeepONet, a specific deep neural operator, and is designed to address several common problems in hydrology, namely, model parameter estimation, prediction at ungaged locations, and uncertainty quantification. Here we demonstrate the effectiveness of our DeepONet‐based workflow using an existing large model ensemble created for an eastern U.S. watershed that is instrumented with 10 streamflow gages. Results suggest DeepONet achieves high efficiency in learning an ML surrogate model from the model ensemble, with the modified Kling‐Gupta Efficiency exceeding 0.9 on holdout test sets. Parameter inference, carried out using the trained DeepONet surrogate model and genetic algorithm, also yields robust results. Additionally, we formulate and train a separate DeepONet model for physics‐informed, seq‐to‐seq streamflow forecasting, which further reduces biases in the pre‐trained DeepONet surrogate model. While this study focuses primarily on a single watershed, our approach is general and may be extended to enable learning from model ensembles across multiple basins or models. Thus, this research represents a significant contribution to the application of hybrid machine learning in hydrology.

Bayesian Inference for Estimating Heat Sources through Temperature Assimilation

Abstract This paper utilizes a Bayesian inference framework to address the two-dimensional steady-state heat conduction problem, focusing on the estimation of unknown distributed heat sources in a thermally-conducting medium with uniform conductivity. The goal is to infer the locations, strength, and shape of heaters by assimilating temperature data in Euclidean space, employing a Fourier series to represent each heater's shape. The Markov Chain Monte Carlo method, incorporating the random-walk Metropolis-Hasting algorithm and parallel tempering, is utilized for posterior distribution exploration in both unbounded and wall-bounded domains. It is found that multiple solutions arise in cases where the number of temperature sensors is less than the number of unknown states. Moreover, smaller heaters introduce greater uncertainty in estimated strength. To address the challenge of estimating the heater's strength and shape simultaneously due to their strong correlation, our method incorporates sharp priors on one to ensure accurate and feasible solutions of the other. The diffusive nature of heat conduction smooths out any deformations in the temperature contours, especially in the presence of multiple heaters positioned near each other, impacting convergence. In wall-bounded domains with Neumann boundary conditions, the inference of heater parameters tends to be more accurate than in unbounded domains.

Analysis of dependent complementary competing risks data from a generalized inverted family of lifetime distributions under a maximum ranked set sampling procedure with unequal samples

This paper explores analysis of a dependent complementary competing risks model when the failure causes are distributed by the proposed generalized inverted family of lifetime distributions. Under maximum ranked set sampling with unequal samples (MRSSU), statistical inference of model parameters and reliability indices is discussed under classical frequentist and Bayesian approaches, respectively. Maximum likelihood estimators along with their existence and uniqueness are obtained for model parameters, and associated approximate confidence intervals are constructed in consequence. Bayesian estimation is also performed with respect to general flexible priors, and the Markov Chain Monte Carlo (MCMC) algorithm is proposed for complex posterior computation. The study further examines classical and Bayesian estimations with order restriction of parameters when additional historical information is available in the MRSSU scenario. Finally, the performance of different results is evaluated through numerical simulations and a real data example is presented for demonstrating the application of our methods.

Efficient approximate Bayesian inference for quantifying uncertainty in multiscale animal movement models

It is becoming increasingly important for wildlife managers and conservation ecologists to understand which resources are selected or avoided by an animal and how to best predict future spatial distributions of animal populations in the long term. However, inferring the patterns of space use by animals is a challenging multiscale inference problem, and formal uncertainty quantification of parameter estimates is an essential component of models that provide useful predictions across scales. In this study, we develop an approximate Bayesian inference framework for step selection models of animal movement which quantifies the uncertainty in estimates of resource selection and avoidance parameters within the Bayesian paradigm. The framework allows joint inference of movement and resource selection parameters of animals and is multiscale in that parameters inferred from fine scale movement steps scale to produce predictions of long-term patterns of space use. Our analysis focuses on simulated movement data in which we test the performance of our framework by altering movement parameters in the data-generating process. In our simulations, individuals respond to two environmental covariates and we employ all combinations of positive and negative selection coefficients corresponding to attraction to an environmental feature and avoidance of an environmental feature, respectively. In all scenarios, we recover the movement parameters used for the simulation of synthetic movement data using variational inference, an approximate Bayesian method, allowing us to formally quantify the uncertainty associated with each parameter for varying data set sizes. Our framework successfully recovered all combinations of movement parameters of the simulated data and accurately captured their posterior distributions given the available data suggesting that the framework is reliable and suitable for inferring how animals select resources and move on a landscape.Notably, our analysis shows that even for reasonably large data sets (circa 10,000 observations) there can still be considerable uncertainty associated with resource selection parameters which can in turn lead to inaccurate predictions of long term space use if not properly incorporated into the modelling approach. To further illustrate the utility of our approach, we also present a case study of its application to an example data set consisting of GPS locations of a fisher (Martes pennanti) Our approach will be of interest to ecologists looking to address conservation questions such as when and where animals are likely to spend most of their time. Furthermore, the approach could be used to predict new suitable areas for conservation based on how GPS collared animal use or avoid resources while including uncertainty around the predictions, thereby helping to make informed management decisions.

Towards faster yet accurate video prediction for resource-constrained platforms

Video prediction has emerged as a critical task in the field of computer vision. While conventional deep learning-based video prediction models excel in accuracy, they often impose significant computational overhead. Moreover, they tend to disregard crucial factors like inference latency and memory consumption, which may render them unsuitable for real-time video prediction applications. This issue is further exacerbated by the necessity to deploy many of these applications on resource-limited embedded devices. In this study, we introduce an innovative faster yet accurate video prediction (FAVP) model designed to mitigate inference latency and memory footprint, all the while maintaining predictive accuracy. Our model employs an encoder-convertor-decoder architecture with a novel multi-layer convertor. Each layer within the convertor uses consistent kernel sizes to precisely capture dimensional features, while varying kernel sizes between layers effectively capture both local and global features. This combination of uniformity within layers and diversity between layers enhances the model’s ability to grasp dynamic changes. Furthermore, we demonstrate that replacing conventional large-kernel convolutions with involutions significantly trims the model’s parameter count and inference latency, without compromising prediction accuracy. Through comprehensive experiments conducted on resource-adequate x86 platforms, utilizing KTH, TrafficBJ, Human3.6M, Moving MNIST, and KTH-Enhanced datasets, we showcase our model’s exceptional prediction performance, particularly in terms of inference latency. Additionally, we evaluate the model on resource-constrained NVIDIA Jetson Nano platform using KITTI & CalTech Pedestrian dataset. The results underscore our model’s superior inference speed and its competence in meeting real-time video prediction requirements, even under stringent resource limitations.

Toward Accelerated Nuclear-physics Parameter Estimation from Binary Neutron Star Mergers: Emulators for the Tolman–Oppenheimer–Volkoff Equations

Gravitational-wave observations of binary neutron-star (BNS) mergers have the potential to revolutionize our understanding of the nuclear equation of state (EOS) and the fundamental interactions that determine its properties. However, Bayesian parameter estimation frameworks do not typically sample over microscopic nuclear-physics parameters that determine the EOS. One of the major hurdles in doing so is the computational cost involved in solving the neutron-star structure equations, known as the Tolman–Oppenheimer–Volkoff (TOV) equations. In this paper, we explore approaches to emulating solutions for the TOV equations: multilayer perceptrons (MLPs), Gaussian processes, and a data-driven variant of the reduced basis method (RBM). We implement these emulators for three different parameterizations of the nuclear EOS, each with a different degree of complexity represented by the number of model parameters. We find that our MLP-based emulators are generally more accurate than the other two algorithms, whereas the RBM results in the largest speedup with respect to the full high-fidelity TOV solver. We employ these emulators for a simple parameter inference using a potentially loud BNS observation and show that the posteriors predicted by our emulators are in excellent agreement with those obtained from the full TOV solver.

Robust preference for Dynamical Dark Energy in DESI BAO and SN measurements

Recent Baryon Acoustic Oscillation (BAO) measurements released by DESI, when combined with Cosmic Microwave Background (CMB) data from Planck and two different samples of Type Ia supernovae (Pantheon-Plus and DESY5) reveal a preference for Dynamical Dark Energy (DDE) characterized by a present-day quintessence-like equation of state that crossed into the phantom regime in the past. A core ansatz for this result is assuming a linear Chevallier-Polarski-Linder (CPL) parameterization w(a) = w 0 + wa (1-a) to describe the evolution of the DE equation of state (EoS). In this paper, we test if and to what extent this assumption impacts the results. To prevent broadening uncertainties in cosmological parameter inference and facilitate direct comparison with the baseline CPL case, we focus on 4 alternative well-known models that, just like CPL, consist of only two free parameters: the present-day DE EoS (w0 ) and a parameter quantifying its dynamical evolution (wa ). We demonstrate that the preference for DDE remains robust regardless of the parameterization: w0 consistently remains in the quintessence regime, while wa consistently indicates a preference for a dynamical evolution towards the phantom regime. This tendency is significantly strengthened by DESY5 SN measurements. By comparing the best-fit χ2 obtained within each DDE model, we notice that the linear CPL parameterization is not the best-fitting case. Among the models considered, the EoS proposed by Barboza and Alcaniz consistently leads to the most significant improvement.

Reionization Parameter Inference from 3D Minkowski Functionals of the 21 cm Signals

The Minkowski functionals (MFs), a set of topological summary statistics, have emerged as a powerful tool for extracting non-Gaussian information. We investigate the prospect of constraining the reionization parameters using the MFs of the 21 cm brightness temperature field from the epoch of reionization (EOR). Realistic effects, including thermal noise, synthesized beam, and foreground avoidance, are applied to the mock observations from radio interferometric array experiments such as the Hydrogen Epoch of Reionization Array (HERA) and the Square Kilometre Array (SKA). We demonstrate that the MFs of the 21 cm signal measured with SKA-Low can be used to distinguish different reionization models, whereas the MF measurement with a HERA-like array cannot be made accurately enough. We further forecast the accuracies with which the MF measurements can place constraints on reionization parameters, using the standard Markov Chain Monte Carlo analysis for parameter inference based on forward modeling. We find that for SKA-Low observation, MFs provide unbiased estimations of the reionization parameters with accuracies comparable to the power spectrum (PS) analysis. Furthermore, joint constraints using both MFs and PS can improve the constraint accuracies by up to 30% compared to those with the PS alone. Nevertheless, the constraint accuracies can be degraded if the EOR window is shrunk with strong foreground avoidance. Our analysis demonstrates the promise of MFs as a set of summary statistics that extract complementary information from the 21 cm EOR field to the two-point statistics, which suggests a strong motivation for incorporating the MFs into the data analysis of future 21 cm observations.

Bayesian parameter inference for epithelial mechanics

Cell-based mechanical models, such as the Cell Vertex Model (CVM), have proven useful for studying the mechanical control of epithelial tissue dynamics. We recently developed a statistical method called image-based parameter inference for formulating CVM model functions and estimating their parameters from image data of epithelial tissues. In this study, we employed Bayesian statistics to improve the utility and flexibility of image-based parameter inference. Tests on synthetic data confirmed that both our non-hierarchical and hierarchical Bayesian models provide accurate estimates of model parameters. By applying this method to Drosophila wings, we demonstrated that the reliability of parameter estimation is closely linked to the mechanical anisotropies present in the tissue. Moreover, we revealed that the cortical elasticity term is dispensable for explaining force-shape correlations in vivo. We anticipate that the flexibility of the Bayesian statistical framework will facilitate the integration of various types of information, thereby contributing to the quantitative dissection of the mechanical control of tissue dynamics.

An evaluation of source-blending impact on the calibration of SKA EoR experiments

ABSTRACT Twenty-one-centimetre signals from the Epoch of Reionization (EoR) are expected to be detected in the low-frequency radio window by the next-generation interferometers, particularly the Square Kilometre Array (SKA). However, precision data analysis pipelines are required to minimize the systematics within an infinitesimal error budget. Consequently, there is a growing need to characterize the sources of errors in EoR analysis. In this study, we identify one such error origin, namely source blending, which is introduced by the overlap of objects in the densely populated observing sky under SKA1-Low’s unprecedented sensitivity and resolution, and evaluate its two-fold impact in both the spatial and frequency domains using a novel hybrid evaluation (HEVAL) pipeline combining end-to-end simulation with an analytic method to mimic EoR analysis pipelines. Sky models corrupted by source blending induce small but severe frequency-dependent calibration errors when coupled with astronomical foregrounds, impeding EoR parameter inference with strong additive residuals in the two-dimensional power spectrum space. We report that additive residuals from poor calibration against sky models with blending ratios of 5 and 0.5 per cent significantly contaminate the EoR window. In contrast, the sky model with a 0.05 per cent blending ratio leaves little residual imprint within the EoR window, therefore identifying a blending tolerance at approximately 0.05 per cent. Given that the SKA observing sky is estimated to suffer from an extended level of blending, strategies involving de-blending, frequency-dependent error mitigation, or a combination of both, are required to effectively attenuate the calibration impact of source-blending defects.

Explanation for the Absence of Secondary Peaks in Black Hole Light Curve Autocorrelations.

The observed radiation from hot gas accreting onto a black hole depends on both the details of the flow and the spacetime geometry. The lensing behavior of a black hole produces a distinctive pattern of autocorrelations within its photon ring that encodes its mass, spin, and inclination. In particular, the time autocorrelation of the light curve is expected to display a series of peaks produced by light echoes of the source, with each peak delayed by the characteristic time lapse τ between light echoes. However, such peaks are absent from the light curves of observed black holes. Here, we develop an analytical model for such light curves that demonstrates how, even though light echoes always exist in the signal, they do not produce autocorrelation peaks if the characteristic correlation timescale λ_{0} of the source is greater than τ. We validate our model against simulated light curves of a stochastic accretion model ray traced with a general-relativistic code, and then fit the model to an observed light curve for Sgr A^{*}. We infer that λ_{0}>τ, providing an explanation for the absence of light echoes in the time autocorrelations of Sgr A^{*} light curves. Our results highlight the importance for black hole parameter inference of spatially resolving the photon ring via future space-based interferometry.

Inferring Binary Properties from Gravitational-Wave Signals

This review provides a conceptual and technical survey of methods for parameter estimation of gravitational-wave signals in ground-based interferometers such as Laser Interferometer Gravitational-Wave Observatory (LIGO) and Virgo. We introduce the framework of Bayesian inference and provide an overview of models for the generation and detection of gravitational waves from compact binary mergers, focusing on the essential features that are observable in the signals. Within the traditional likelihood-based paradigm, we describe various approaches for enhancing the efficiency and robustness of parameter inference. This includes techniques for accelerating likelihood evaluations, such as heterodyne/relative binning, reduced-order quadrature, multibanding, and interpolation. We also cover methods to simplify the analysis to improve convergence, via reparameterization, importance sampling, and marginalization. We end with a discussion of recent developments in the application of likelihood-free (simulation-based) inference methods to gravitational-wave data analysis.

Parallel Sampling of Decomposable Graphs Using Markov Chains on Junction Trees

Bayesian inference for undirected graphical models is mostly restricted to the class of decomposable graphs, as they enjoy a rich set of properties making them amenable to high-dimensional problems. While parameter inference is straightforward in this setup, inferring the underlying graph is a challenge driven by the computational difficulty in exploring the space of decomposable graphs. This work makes two contributions to address this problem. First, we provide sufficient and necessary conditions for when multi-edge perturbations maintain decomposability of the graph. Using these, we characterize a simple class of partitions that efficiently classify all edge perturbations by whether they maintain decomposability. Second, we propose a novel parallel nonreversible Markov chain Monte Carlo sampler for distributions over junction tree representations of the graph. At every step, the parallel sampler executes simultaneously all edge perturbations within a partition. Through simulations, we demonstrate the efficiency of our new edge perturbation conditions and class of partitions. We find that our parallel sampler yields improved mixing properties in comparison to the single-move variate, and outperforms current state-of-the-art methods in terms of accuracy and computational efficiency. The implementation of our work is available in the Python package parallelDG. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Model simplification to simulate groundwater recharge from a perched gravel-bed river

Gravel-bed rivers are an important source of groundwater recharge in some regions of the world. Their interactions with groundwater are complex and highly variable in space and time, with considerable water storage in the riverbed sediments. In losing river sections, where most of the groundwater recharge occurs, the river can be separated from the regional groundwater system by an unsaturated zone (i.e., perched). The complexity of groundwater–surface water interactions in these environments calls for the use of 3D fully integrated hydrological models to represent them, but their computational intensity limits their practicality for parameter inference, uncertainty quantification and regional scale problems. On the other hand, the simple groundwater–surface water exchange functions currently implemented in regional scale groundwater models are not suited to represent complex gravel-bed river systems such as braided rivers. There is therefore a need for developing groundwater–surface water exchange functions tailored to gravel-bed rivers that can be used in regional scale models.To address this issue, we developed a model simplification framework that combines a 3D integrated surface and subsurface hydrological model, a 2D cross-sectional river-aquifer model and a 1D conductance-based analytical model. We aim at broadly simplifying the 3D model while ensuring the appropriate simulation of groundwater recharge. We demonstrate our modelling approach on the Selwyn River (New Zealand) using piezometric data and groundwater recharge estimates derived from field observations and satellite imagery.Our results indicates that groundwater recharge from this river can be simulated using a simple 1D analytical model, which can easily be implemented in regional groundwater models (e.g., MODFLOW models). However, to represent properly the time variability of groundwater recharge, it is essential to use the groundwater level in the shallow aquifer associated with the river as input to the regional groundwater model. Our approach is generally transferable to other gravel-bed rivers but requires some observations of river losses for proper calibration.

Model‐based clustering in simple hypergraphs through a stochastic blockmodel

AbstractWe propose a model to address the overlooked problem of node clustering in simple hypergraphs. Simple hypergraphs are suitable when a node may not appear multiple times in the same hyperedge, such as in co‐authorship datasets. Our model generalizes the stochastic blockmodel for graphs and assumes the existence of latent node groups and hyperedges are conditionally independent given these groups. We first establish the generic identifiability of the model parameters. We then develop a variational approximation Expectation‐Maximization algorithm for parameter inference and node clustering, and derive a statistical criterion for model selection. To illustrate the performance of our R package HyperSBM, we compare it with other node clustering methods using synthetic data generated from the model, as well as from a line clustering experiment and a co‐authorship dataset.