arXiv subjects

John V. Tucker

Publications and source records attributed to John V. Tucker.

7 recordsLinked to original sources

arXiv2024

The dynamics of belief: continuously monitoring and visualising complex systems

The rise of AI in human contexts places new demands on automated systems to be transparent and explainable. We examine some anthropomorphic ideas and principles relevant to such accountablity in order to develop a theoretical framework for thinking about digital systems in complex human contexts and the problem of explaining their behaviour. Structurally, systems are made of modular and hierachical components, which we abstract in a new system model using notions of modes and mode transitions. A mode is an independent component of the system with its own objectives, monitoring data, and algorithms. The behaviour of a mode, including its transitions to other modes, is determined by functions that interpret each mode's monitoring data in the light of its objectives and algorithms. We show how these belief functions can help explain system behaviour by visualising their evaluation as trajectories in higher-dimensional geometric spaces. These ideas are formalised mathematically by abstract and concrete simplicial complexes. We offer three techniques: a framework for design heuristics, a general system theory based on modes, and a geometric visualisation, and apply them in three types of human-centred systems.

cs.AI↗

arXiv2021

A model of systems with modes and mode transitions

We propose a method of classifying the operation of a system into finitely many modes. Each mode has its own objectives for the system's behaviour and its own mathematical models and algorithms designed to accomplish its objectives. A central problem is deciding when to transition from one mode to some other mode, a decision that may be contested and involve partial or inconsistent information or evidence. We model formally the concept of modes for a system and derive a family of data types for analysing mode transitions. The data types are simplicial complexes, both abstract and realised in euclidean space $\mathbb{R}^{n}$. In the data type, a mode is represented by a simplex. Each state of a system can be evaluated relative to different modes by mapping it into one or more simplices. This calibration measures the extent to which distinct modes are appropriate for the state and can decide on a transition. We explain this methodology based on modes, introduce the mathematical ideas about simplicial objects we need and use them to build a theoretical framework for modes and mode transitions. To illustrate the general model in some detail, we work though a case study of an autonomous racing car.

cs.LO↗

arXiv2018

An algebraic theory for data linkage

There are countless sources of data available to governments, companies, and citizens, which can be combined for good or evil. We analyse the concepts of combining data from common sources and linking data from different sources. We model the data and its information content to be found in a single source by a partial ordered monoid, and the transfer of information between sources by different types of morphisms. To capture the linkage between a family of sources, we use a form of Grothendieck construction to create a partial ordered monoid that brings together the global data of the family in a single structure. We apply our approach to database theory and axiomatic structures in approximate reasoning. Thus, partial ordered monoids provide a foundation for the algebraic study for information gathering in its most primitive form.

cs.LO↗

arXiv2017

Monitoring and Intervention: Concepts and Formal Models

Our machines, products, utilities, and environments have long been monitored by embedded software systems. Our professional, commercial, social and personal lives are also subject to monitoring as they are mediated by software systems. Data on nearly everything now exists, waiting to be collected and analysed for all sorts of reasons. Given the rising tide of data we pose the questions: What is monitoring? Do diverse and disparate monitoring systems have anything in common? We attempt answer these questions by proposing an abstract conceptual framework for studying monitoring. We argue that it captures a structure common to many different monitoring practices, and that from it detailed formal models can be derived, customised to applications. The framework formalises the idea that monitoring is a process that observes the behaviour of people and objects in a context. The entities and their behaviours are represented by abstract data types and the observable attributes by logics. Since monitoring usually has a specific purpose, we extend the framework with protocols for detecting attributes or events that require interventions and, possibly, a change in behaviour. Our theory is illustrated by a case study from criminal justice, that of electronic tagging.

cs.CY↗

arXiv2016

Analogue-digital systems with modes of physical behaviour

Complex environments, processes and systems may exhibit several distinct modes of physical behaviour or operation. Thus, for example, in their design, a set of mathematical models may be needed, each model having its own domain of application and representing a particular mode of behaviour or operation of physical reality. The models may be of disparate kinds { discrete or continuous in data, time and space. Furthermore, some physical modes may not have a reliable model. Physical measurements determine modes of operation. We explore the question: What is a mode of behaviour? How do we specify algorithms and software that monitor or govern a complex physical situation with many modes? How do we specify a portfolio of modes, and the computational problem of transitioning from using one mode to another mode as physical modes change? We propose a general definition of an analogue-digital system with modes. We show how any diverse set of modes { with or without models { can be bound together, and how the transitions between modes can be determined, by constructing a topological data type based upon a simplicial complex. We illustrate the ideas of physical modes and our theory by reflecting on simple examples, including driverless racing cars.

cs.SE↗

arXiv2014

On the Role of Identity in Surveillance

Surveillance is a process that observes behaviour, recognises properties and identifies individuals. It has become a commonplace phenomenon in our everyday life. Many surveillance practices depend on the use of advanced technologies to collect, store and process data. We propose (i) an abstract definition of surveillance; and (ii) an abstract definition of identity, designed to capture the common structure of many disparate surveillance situations. We argue that the notion of identity is fundamental to surveillance. Rather than having a single identity, individuals have many identities, real and virtual, that are used in different aspects of their lives. Most aspects of life are subject to some form of surveillance, and observations and identities can be aggregated. The notion of identity needs to be theorised. Our analysis is very general and, at the same time, sufficiently precise to be the basis of mathematical models.

cs.CY↗

arXiv2014

Formalising Surveillance and Identity

Surveillance is a social phenomenon that is general and commonplace, employed by governments, companies and communities. Its ubiquity is due to technologies for gathering and processing data; its strong and obvious effects raise difficult social questions. We give a general definition of surveillance that captures the notion in diverse situations and we illustrate it with some disparate examples.A most important, if neglected,component idea is that of the identity of the people or objects observed. We propose a general definition of identifiers as data designed to specify the identity of an entity in some context or for some purpose. We examine the ways identifiers depend upon other identifiers and show the provenance of identifiers requires reductions between identifiers and a special idea of personal identifier. The theory is formalised mathematically. Finally, we reflect on the role of formal methods to give insights in sociological contexts.

cs.CY↗