arXiv subjects

Jim Q. Smith

Publications and source records attributed to Jim Q. Smith.

At least 37 records · Page 2Linked to original sources

arXiv2020

Propagation for Dynamic Continuous Time Chain Event Graphs

Chain Event Graphs (CEGs) are a family of event-based graphical models that represent context-specific conditional independences typically exhibited by asymmetric state space problems. The class of continuous time dynamic CEGs (CT-DCEGs) provides a factored representation of longitudinally evolving trajectories of a process in continuous time. Temporal evidence in a CT-DCEG introduces dependence between its transition and holding time distributions. We present a tractable exact inferential scheme analogous to the scheme in Kjærulff (1992) for discrete Dynamic Bayesian Networks (DBNs) which employs standard junction tree inference by "unrolling" the DBN. To enable this scheme, we present an extension of the standard CEG propagation algorithm (Thwaites et al., 2008). Interestingly, the CT-DCEG benefits from simplification of its graph on observing compatible evidence while preserving the still relevant symmetries within the asymmetric network. Our results indicate that the CT-DCEG is preferred to DBNs and continuous time BNs under contexts involving significant asymmetry and a natural total ordering of the process evolution.

cs.AI↗

arXiv2020

A decision support system for addressing food security in the UK

This paper presents an integrating decision support system to model food security in the UK. In ever-larger dynamic systems, such as the food system, it is increasingly difficult for decision-makers to effectively account for all the variables within the system that may influence the outcomes of interest under enactments of various candidate policies. Each of the influencing variables is likely, themselves, to be dynamic sub-systems with expert domains supported by sophisticated probabilistic models. Recent increases in food poverty the UK raised the questions about the main drivers to food insecurity, how this may be changing over time and how evidence can be used in evaluating policy for decision support. In this context, an integrating decision support system is proposed for household food security to allow decision-makers to compare several candidate policies which may affect the outcome of food insecurity at household level.

stat.AP↗

arXiv2020

Bayesian Learning of Causal Relationships for System Reliability

Causal theory is now widely developed with many applications to medicine and public health. However within the discipline of reliability, although causation is a key concept in this field, there has been much less theoretical attention. In this paper, we will demonstrate how some aspects of established causal methodology can be translated via trees, and more specifically chain event graphs, into domain of reliability theory to help the probability modeling of failures. We further show how various domain specific concepts of causality particular to reliability can be imported into more generic causal algebras and so demonstrate how these disciplines can inform each other. This paper is informed by a detailed analysis of maintenance records associated with a large electrical distribution company. Causal hypotheses embedded within these natural language texts are extracted and analyzed using the new graphical framework we introduced here.

cs.AI↗

arXiv2019

The Scenario Culture

Scenario Analysis is a risk assessment tool that aims to evaluate the impact of a small number of distinct plausible future scenarios. In this paper, we provide an overview of important aspects of Scenario Analysis including when it is appropriate, the design of scenarios, uncertainty and encouraging creativity. Each of these issues is discussed in the context of climate, energy and legal scenarios.

stat.OT↗

arXiv2019

A Bayesian Dynamic Graphical Model for Recurrent Events in Public Health

To analyze the impacts of certain types of public health interventions we need to estimate the treatment effects and outcomes as these apply to heterogeneous open populations. Dynamically modifying populations containing risk groups that can react very differently to changes in covariates are inferentially challenging. Here we propose a novel Bayesian graphical model called the Reduced Dynamic Chain Event Graph (RDCEG) customized to such populations. These models generalize the tree-based Chain Event Graphs to a particular class of graphically supported semi-Markov processes. They provide an interface between natural language explanations about what might be happening to individuals and a formal statistical analysis. Here we show how the RDCEG is able to express the different possible progressions of each vulnerable individual as well as hypotheses about probabilistic symmetries within these progressions across different individuals within that population. We demonstrate how well-developed Bayesian Network technologies can be transferred almost seamlessly to this class. Our work is motivated by the challenge of modeling non-pharmacological interventions for recurrent event processes. We illustrate our methodology in two settings: an intervention to reduce falls among the elderly and a trial to examine effects of deferred treatment among individuals presenting with early epilepsy.

stat.ME↗

arXiv2019

Bayesian Diagnostics for Chain Event Graphs

Chain event graphs have been established as a practical Bayesian graphical tool. While bespoke diagnostics have been developed for Bayesian Networks, they have not yet been defined for the statistical class of Chain Event Graph models. Mirroring the methodology of prequential diagnostics for Bayesian Networks, in this paper we design a number of new Bayesian diagnostics for this new class. These can be used to check whether a selected model--presumably the best within the class--captures most of the salient features of the observed process. These are designed to check the continued validity of a selected model as data about a population is collected. A previous study of childhood illness in New Zealand illustrates the efficacy of these diagnostics. A second example on radicalisation is used as a more expressive example.

stat.ME↗

arXiv2019

GASC: Genre-Aware Semantic Change for Ancient Greek

Word meaning changes over time, depending on linguistic and extra-linguistic factors. Associating a word's correct meaning in its historical context is a central challenge in diachronic research, and is relevant to a range of NLP tasks, including information retrieval and semantic search in historical texts. Bayesian models for semantic change have emerged as a powerful tool to address this challenge, providing explicit and interpretable representations of semantic change phenomena. However, while corpora typically come with rich metadata, existing models are limited by their inability to exploit contextual information (such as text genre) beyond the document time-stamp. This is particularly critical in the case of ancient languages, where lack of data and long diachronic span make it harder to draw a clear distinction between polysemy (the fact that a word has several senses) and semantic change (the process of acquiring, losing, or changing senses), and current systems perform poorly on these languages. We develop GASC, a dynamic semantic change model that leverages categorical metadata about the texts' genre to boost inference and uncover the evolution of meanings in Ancient Greek corpora. In a new evaluation framework, our model achieves improved predictive performance compared to the state of the art.

cs.CL↗

arXiv2018

An N Time-Slice Dynamic Chain Event Graph

The Dynamic Chain Event Graph (DCEG) is able to depict many classes of discrete random processes exhibiting asymmetries in their developments and context-specific conditional probabilities structures. However, paradoxically, this very generality has so far frustrated its wide application. So in this paper we develop an object-oriented method to fully analyse a particularly useful and feasibly implementable new subclass of these graphical models called the N Time-Slice DCEG (NT-DCEG). After demonstrating a close relationship between an NT-DCEG and a specific class of Markov processes, we discuss how graphical modellers can exploit this connection to gain a deep understanding of their processes. We also show how to read from the topology of this graph context-specific independence statements that can then be checked by domain experts. Our methods are illustrated throughout using examples of dynamic multivariate processes describing inmate radicalisation in a prison.

stat.ML↗

arXiv2018

Bayesian Networks, Total Variation and Robustness

Now that Bayesian Networks (BNs) have become widely used, an appreciation is developing of just how critical an awareness of the sensitivity and robustness of certain target variables are to changes in the model. When time resources are limited, such issues impact directly on the chosen level of complexity of the BN as well as the quantity of missing probabilities we are able to elicit. Currently most such analyses are performed once the whole BN has been elicited and are based on Kullback-Leibler information measures. In this paper we argue that robustness methods based instead on the familiar total variation distance provide simple and more useful bounds on robustness to misspecification which are both formally justifiable and transparent. We demonstrate how such formal robustness considerations can be embedded within the process of building a BN. Here we focus on two particular choices a modeller needs to make: the choice of the parents of each node and the number of levels to choose for each variable within the system. Our analyses are illustrated throughout using two BNs drawn from the recent literature.

stat.ME↗

arXiv2018

Properties of an N Time-Slice Dynamic Chain Event Graph

A Dynamic Chain Event Graph (DCEG) provides a rich tree-based framework for modelling a dynamic process with highly asymmetric developments. An N Time-Slice DCEG (NT-DCEG) is a useful subclass of the DCEG class that exhibits a specific type of periodicity in its supporting tree graph and embodies a time-homogeneity assumption. Here some desired properties of an NT-DCEG is explored. In particular, we prove that the class of NT-DCEGs contains all discrete N time-slice Dynamic Bayesian Networks as special cases. We also develop a method to distributively construct an NT-DCEG model. By exploiting the topology of an NT-DCEG graph, we show how to construct intrinsic random variables which exhibit context-specific independences that can then be checked by domain experts. We also show how an NT-DCEG can be used to depict various structural and Granger causal hypotheses about a given process. Our methods are illustrated throughout using examples of dynamic multivariate processes describing inmate radicalisation in a prison.

stat.ML↗

arXiv2018

A conditional independence framework for coherent modularized inference

Inference in current domains of application are often complex and require us to integrate the expertise of a variety of disparate panels of experts and models coherently. In this paper we develop a formal statistical methodology to guide the networking together of a diverse collection of probabilistic models. In particular, we derive sufficient conditions that ensure inference remains coherent across the composite before and after accommodating relevant evidence.

stat.ME↗

arXiv2018

Customised Structural Elicitation

Established methods for structural elicitation typically rely on code modelling standard graphical models classes, most often Bayesian networks. However, more appropriate models may arise from asking the expert questions in common language about what might relate to what and exploring the logical implications of the statements. Only after identifying the best matching structure should this be embellished into a fully quantified probability model. Examples of the efficacy and potential of this more flexible approach are shown below for four classes of graphical models: Bayesian networks, Chain Event Graphs, Multi-regression Dynamic Models, and Flow Graphs. We argue that to be fully effective any structural elicitation phase must first be customised to an application and if necessary new types of structure with their own bespoke semantics elicited.

stat.ME↗

arXiv2017

Coherent combination of probabilistic outputs for group decision making: an algebraic approach

Current decision support systems address domains that are heterogeneous in nature and becoming progressively larger. Such systems often require the input of expert judgement about a variety of different fields and an intensive computational power to produce the scores necessary to rank the available policies. Recently, integrating decision support systems have been introduced to enable a formal Bayesian multi-agent decision analysis to be distributed and consequently efficient. In such systems, where different panels of experts oversee disjoint but correlated vectors of variables, each expert group needs to deliver only certain summaries of the variables under their jurisdiction to properly derive an overall score for the available policies. Here we present an algebraic approach that makes this methodology feasible for a wide range of modelling contexts and that enables us to identify the summaries needed for such a combination of judgements. We are also able to demonstrate that coherence, in a sense we formalize here, is still guaranteed when panels only share a partial specification of their model with other panel members. We illustrate this algebraic approach by applying it to a specific class of Bayesian networks and demonstrate how we can use it to derive closed form formulae for the computations of the joint moments of variables that determine the score of different policies.

stat.OT↗

arXiv2017

Equivalence Classes of Staged Trees

In this paper we give a complete characterization of the statistical equivalence classes of CEGs and of staged trees. We are able to show that all graphical representations of the same model share a common polynomial description. Then, simple transformations on that polynomial enable us to traverse the corresponding class of graphs. We illustrate our results with a real analysis of the implicit dependence relationships within a previously studied dataset.

math.ST↗

arXiv2017

Discovery of statistical equivalence classes using computer algebra

Discrete statistical models supported on labelled event trees can be specified using so-called interpolating polynomials which are generalizations of generating functions. These admit a nested representation. A new algorithm exploits the primary decomposition of monomial ideals associated with an interpolating polynomial to quickly compute all nested representations of that polynomial. It hereby determines an important subclass of all trees representing the same statistical model. To illustrate this method we analyze the full polynomial equivalence class of a staged tree representing the best fitting model inferred from a real-world dataset.

math.ST↗

arXiv2017

A Graphical method for simplifying Bayesian Games

If the influence diagram (ID) depicting a Bayesian game is common knowledge to its players then additional assumptions may allow the players to make use of its embodied irrelevance statements. They can then use these to discover a simpler game which still embodies both their optimal decision policies. However the impact of this result has been rather limited because many common Bayesian games do not exhibit sufficient symmetry to be fully and efficiently represented by an ID. The tree-based chain event graph (CEG) has been developed specifically for such asymmetric problems. By using these graphs rational players can make analogous deductions, assuming the topology of the CEG as common knowledge. In this paper we describe these powerful new techniques and illustrate them through an example modelling a game played between a government department and the provider of a website designed to radicalise vulnerable people.

cs.GT↗

arXiv2017

A symbolic algebra for the computation of expected utilities in multiplicative influence diagrams

Influence diagrams provide a compact graphical representation of decision problems. Several algorithms for the quick computation of their associated expected utilities are available in the literature. However, often they rely on a full quantification of both probabilistic uncertainties and utility values. For problems where all random variables and decision spaces are finite and discrete, here we develop a symbolic way to calculate the expected utilities of influence diagrams that does not require a full numerical representation. Within this approach expected utilities correspond to families of polynomials. After characterizing their polynomial structure, we develop an efficient symbolic algorithm for the propagation of expected utilities through the diagram and provide an implementation of this algorithm using a computer algebra system. We then characterize many of the standard manipulations of influence diagrams as transformations of polynomials. We also generalize the decision analytic framework of these diagrams by defining asymmetries as operations over the expected utility polynomials.

cs.AI↗

arXiv2016

Directed expected utility networks

A variety of statistical graphical models have been defined to represent the conditional independences underlying a random vector of interest. Similarly, many different graphs embedding various types of preferential independences, as for example conditional utility independence and generalized additive independence, have more recently started to appear. In this paper we define a new graphical model, called a directed expected utility network, whose edges depict both probabilistic and utility conditional independences. These embed a very flexible class of utility models, much larger than those usually conceived in standard influence diagrams. Our graphical representation, and various transformations of the original graph into a tree structure, are then used to guide fast routines for the computation of a decision problem's expected utilities. We show that our routines generalize those usually utilized in standard influence diagrams' evaluations under much more restrictive conditions. We then proceed with the construction of a directed expected utility network to support decision makers in the domain of household food security.

cs.AI↗