arXiv subjects

Scott Ferson

Publications and source records attributed to Scott Ferson.

10 recordsLinked to original sources

arXiv2026

Conformalised imprecise inference for robust extrapolation under limited data

Recent advances in uncertainty quantification increasingly emphasise the distinction between aleatory and epistemic uncertainty in machine learning, motivating the need for more unified frameworks. However, despite much progress in producing reliable predictions, existing methods often lack rigorous guarantees when generalising beyond the training domain. We propose a conformalised imprecise inference framework for robust extrapolation, which is model-agnostic and augments predictive models with imprecision and distance awareness. The proposed approach yields imprecise predictions (probability boxes) that remain valid under distributional shift, maintaining coverage while adaptively expanding uncertainty in extrapolation regimes. Experiments on synthetic and benchmark datasets demonstrate improved robustness and reliable coverage compared to standard probabilistic approaches, particularly under limited data.

cs.LG↗

arXiv2023

Correlation-Based And-Operations Can Be Copulas: A Proof

In many practical situations, we know the probabilities $a$ and $b$ of two events $A$ and $B$, and we want to estimate the joint probability ${\rm Prob}(A\,\&\,B)$. The algorithm that estimates the joint probability based on the known values $a$ and $b$ is called an and-operation. An important case when such a reconstruction is possible is when we know the correlation between $A$ and $B$; we call the resulting and-operation correlation-based. On the other hand, in statistics, there is a widely used class of and-operations known as copulas. Empirical evidence seems to indicate that the correlation-based and-operation derived in https://doi.org/10.1007/978-3-031-08971-8_64 is a copula, but until now, no proof of this statement was available. In this paper, we provide such a proof.

stat.OT↗

arXiv2022

Should data ever be thrown away? Pooling interval-censored data sets with different precision

Data quality is an important consideration in many engineering applications and projects. Data collection procedures do not always involve careful utilization of the most precise instruments and strictest protocols. As a consequence, data are invariably affected by imprecision and sometimes sharply varying levels of quality of the data. Different mathematical representations of imprecision have been suggested, including a classical approach to censored data which is considered optimal when the proposed error model is correct, and a weaker approach called interval statistics based on partial identification that makes fewer assumptions. Maximizing the quality of statistical results is often crucial to the success of many engineering projects, and a natural question that arises is whether data of differing qualities should be pooled together or we should include only precise measurements and disregard imprecise data. Some worry that combining precise and imprecise measurements can depreciate the overall quality of the pooled data. Some fear that excluding data of lesser precision can increase their overall uncertainty about results because lower sample size implies more sampling uncertainty. This paper explores these concerns and describes simulation results that show when it is advisable to combine fairly precise data with rather imprecise data by comparing analyses using different mathematical representations of imprecision. Pooling data sets is preferred when the low-quality data set does not exceed a certain level of uncertainty. However, so long as the data are random, it may be legitimate to reject the low-quality data if its reduction of sampling uncertainty does not counterbalance the effect of its imprecision on the overall uncertainty.

stat.ME↗

arXiv2022

Correlated Boolean Operators for Uncertainty Logic

We present a correlated \textit{and} gate which may be used to propagate uncertainty and dependence through Boolean functions, since any Boolean function may be expressed as a combination of \textit{and} and \textit{not} operations. We argue that the \textit{and} gate is a bivariate copula family, which has the interpretation of constructing bivariate Bernoulli random variables following a given Pearson correlation coefficient and marginal probabilities. We show how this copula family may be used to propagate uncertainty in the form of probabilities of events, probability intervals, and probability boxes, with only partial or no knowledge of the dependency between events, expressed as an interval for the correlation coefficient. These results generalise previous results by Fr\'echet on the conjunction of two events with unknown dependencies. We show an application propagating uncertainty through a fault tree for a pressure tank. This paper comes with an open-source Julia library for performing uncertainty logic.

math.PR↗

arXiv2022

Neural network model for imprecise regression with interval dependent variables

This paper presents a computationally feasible method to compute rigorous bounds on the interval-generalisation of regression analysis to account for epistemic uncertainty in the output variables. The new iterative method uses machine learning algorithms to fit an imprecise regression model to data that consist of intervals rather than point values. The method is based on a single-layer interval neural network which can be trained to produce an interval prediction. It seeks parameters for the optimal model that minimizes the mean squared error between the actual and predicted interval values of the dependent variable using a first-order gradient-based optimization and interval analysis computations to model the measurement imprecision of the data. An additional extension to a multi-layer neural network is also presented. We consider the explanatory variables to be precise point values, but the measured dependent values are characterized by interval bounds without any probabilistic information. The proposed iterative method estimates the lower and upper bounds of the expectation region, which is an envelope of all possible precise regression lines obtained by ordinary regression analysis based on any configuration of real-valued points from the respective y-intervals and their x-values.

physics.data-an↗

arXiv2021

The Creation of Puffin, the Automatic Uncertainty Compiler

An uncertainty compiler is a tool that automatically translates original computer source code lacking explicit uncertainty analysis into code containing appropriate uncertainty representations and uncertainty propagation algorithms. We have developed an prototype uncertainty compiler along with an associated object-oriented uncertainty language in the form of a stand-alone Python library. It handles the specifications of input uncertainties and inserts calls to intrusive uncertainty quantification algorithms in the library. The uncertainty compiler can apply intrusive uncertainty propagation methods to codes or parts of codes and therefore more comprehensively and flexibly address both epistemic and aleatory uncertainties.

cs.MS↗

arXiv2021

Singhing with Confidence: Visualising the Performance of Confidence Structures

Confidence intervals are an established means of portraying uncertainty about an inferred parameter and can be generated through the use of confidence distributions. For a confidence distribution to be ideal, it must maintain frequentist coverage of the true parameter. This can be represented for a precise distribution by adherence to a cumulative unit uniform distribution, referred to here as a Singh plot. This manuscript extends this to imprecise confidence structures with bounds around the uniform distribution, and describes how deviations convey information regarding the characteristics of confidence structures designed for inference and prediction. This quick visual representation, in a manner similar to ROC curves, aids the development of robust structures and methods that make use of confidence. A demonstration of the utility of Singh plots is provided with an assessment of the coverage of the ProUCL Chebyshev upper confidence limit estimator for the mean of an unknown distribution.

stat.ME↗

arXiv2021

Logistic Regression Through the Veil of Imprecise Data

Logistic regression is an important statistical tool for assessing the probability of an outcome based upon some predictive variables. Standard methods can only deal with precisely known data, however many datasets have uncertainties which traditional methods either reduce to a single point or completely disregarded. In this paper we show that it is possible to include these uncertainties by considering an imprecise logistic regression model using the set of possible models that can be obtained from values from within the intervals. This has the advantage of clearly expressing the epistemic uncertainty removed by traditional methods.

stat.ME↗

arXiv2018

Quantification of Incertitude in Black Box Simulation Codes

We present early results from a study addressing the question of how one treats the propagation of incertitude, that is, epistemic uncertainty, in input parameters in astrophysical simulations. As an example, we look at the propagation of incertitude in control parameters for stellar winds in MESA stellar evolution simulations. We apply two methods of incertitude propagation, the Cauchy Deviates method and the Quadratic Response Surface method, to quantify the output uncertainty in the final white dwarf mass given a range of values for wind parameters. The methodology we apply is applicable to the problem of propagating input incertitudes through any simulation code treated as a "black box," i.e. a code for which the algorithmic details are either inaccessible or prohibitively complicated. We have made the tools developed for this study freely available to the community.

astro-ph.IM↗

arXiv2017

Satellite conjunction analysis and the false confidence theorem

Satellite conjunction analysis is the assessment of collision risk during a close encounter between a satellite and another object in orbit. A counterintuitive phenomenon has emerged in the conjunction analysis literature, namely, probability dilution, in which lower quality data paradoxically appear to reduce the risk of collision. We show that probability dilution is a symptom of a fundamental deficiency in probabilistic representations of statistical inference, in which there are propositions that will consistently be assigned a high degree of belief, regardless of whether or not they are true. We call this deficiency false confidence. In satellite conjunction analysis, it results in a severe and persistent underestimate of collision risk exposure. We introduce the Martin--Liu validity criterion as a benchmark by which to identify statistical methods that are free from false confidence. Such inferences will necessarily be non-probabilistic. In satellite conjunction analysis, we show that uncertainty ellipsoids satisfy the validity criterion. Performing collision avoidance maneuvers based on ellipsoid overlap will ensure that collision risk is capped at the user-specified level. Further, this investigation into satellite conjunction analysis provides a template for recognizing and resolving false confidence issues as they occur in other problems of statistical inference.

math.ST↗