-
On human-centred security: A new systems model based on modes and mode transitions
Authors:
Edwin J Beggs,
John V Tucker,
Victoria Wang
Abstract:
We propose an abstract conceptual framework for analysing complex security systems using a new notion of modes and mode transitions. A mode is an independent component of a system with its own objectives, monitoring data, algorithms, and scope and limits. The behaviour of a mode, including its transitions to other modes, is determined by interpretations of the mode's monitoring data in the light o…
▽ More
We propose an abstract conceptual framework for analysing complex security systems using a new notion of modes and mode transitions. A mode is an independent component of a system with its own objectives, monitoring data, algorithms, and scope and limits. The behaviour of a mode, including its transitions to other modes, is determined by interpretations of the mode's monitoring data in the light of its objectives and capabilities -- these interpretations we call beliefs. We formalise the conceptual framework mathematically and, by quantifying and visualising beliefs in higher-dimensional geometric spaces, we argue our models may help both design, analyse and explain systems. The mathematical models are based on simplicial complexes.
△ Less
Submitted 3 May, 2024;
originally announced May 2024.
-
The serotonergic psychedelic N,N-dipropyltryptamine alters information-processing dynamics in cortical neural circuits
Authors:
Thomas F. Varley,
Daniel Havert,
Leandro Fosque,
Abolfazl Alipour,
Naruepon Weerawongphrom,
Hiroki Naganobori,
Lily O'Shea,
Maria Pope,
John Beggs
Abstract:
Most of the recent work in psychedelic neuroscience has been done using non-invasive neuroimaging, with data recorded from the brains of adult volunteers under the influence of a variety of drugs. While this data provides holistic insights into the effects of psychedelics on whole-brain dynamics, the effects of psychedelics on the meso-scale dynamics of cortical circuits remains much less explored…
▽ More
Most of the recent work in psychedelic neuroscience has been done using non-invasive neuroimaging, with data recorded from the brains of adult volunteers under the influence of a variety of drugs. While this data provides holistic insights into the effects of psychedelics on whole-brain dynamics, the effects of psychedelics on the meso-scale dynamics of cortical circuits remains much less explored. Here, we report the effects of the serotonergic psychedelic N,N-diproptyltryptamine (DPT) on information-processing dynamics in a sample of in vitro organotypic cultures made from rat cortical tissue. Three hours of spontaneous activity were recorded: an hour of pre-drug control, and hour of exposure to 10$μ$M DPT solution, and a final hour of washout, once again under control conditions. We found that DPT reversibly alters information dynamics in multiple ways: first, the DPT condition was associated with higher entropy of spontaneous firing activity and reduced the amount of time information was stored in individual neurons. Second, DPT also reduced the reversibility of neural activity, increasing the entropy produced and suggesting a drive away from equilibrium. Third, DPT altered the structure of neuronal circuits, decreasing the overall information flow coming into each neuron, but increasing the number of weak connections, creating a dynamic that combines elements of integration and disintegration. Finally, DPT decreased the higher-order statistical synergy present in sets of three neurons. Collectively, these results paint a complex picture of how psychedelics regulate information processing in meso-scale cortical tissue. Implications for existing hypotheses of psychedelic action, such as the Entropic Brain Hypothesis, are discussed.
△ Less
Submitted 31 October, 2023;
originally announced October 2023.
-
GRAIMATTER Green Paper: Recommendations for disclosure control of trained Machine Learning (ML) models from Trusted Research Environments (TREs)
Authors:
Emily Jefferson,
James Liley,
Maeve Malone,
Smarti Reel,
Alba Crespi-Boixader,
Xaroula Kerasidou,
Francesco Tava,
Andrew McCarthy,
Richard Preen,
Alberto Blanco-Justicia,
Esma Mansouri-Benssassi,
Josep Domingo-Ferrer,
Jillian Beggs,
Antony Chuter,
Christian Cole,
Felix Ritchie,
Angela Daly,
Simon Rogers,
Jim Smith
Abstract:
TREs are widely, and increasingly used to support statistical analysis of sensitive data across a range of sectors (e.g., health, police, tax and education) as they enable secure and transparent research whilst protecting data confidentiality. There is an increasing desire from academia and industry to train AI models in TREs. The field of AI is developing quickly with applications including spott…
▽ More
TREs are widely, and increasingly used to support statistical analysis of sensitive data across a range of sectors (e.g., health, police, tax and education) as they enable secure and transparent research whilst protecting data confidentiality. There is an increasing desire from academia and industry to train AI models in TREs. The field of AI is developing quickly with applications including spotting human errors, streamlining processes, task automation and decision support. These complex AI models require more information to describe and reproduce, increasing the possibility that sensitive personal data can be inferred from such descriptions. TREs do not have mature processes and controls against these risks. This is a complex topic, and it is unreasonable to expect all TREs to be aware of all risks or that TRE researchers have addressed these risks in AI-specific training. GRAIMATTER has developed a draft set of usable recommendations for TREs to guard against the additional risks when disclosing trained AI models from TREs. The development of these recommendations has been funded by the GRAIMATTER UKRI DARE UK sprint research project. This version of our recommendations was published at the end of the project in September 2022. During the course of the project, we have identified many areas for future investigations to expand and test these recommendations in practice. Therefore, we expect that this document will evolve over time.
△ Less
Submitted 3 November, 2022;
originally announced November 2022.
-
The dynamics of belief: continuously monitoring and visualising complex systems
Authors:
Edwin J. Beggs,
John V. Tucker
Abstract:
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 hierachica…
▽ More
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.
△ Less
Submitted 17 January, 2024; v1 submitted 11 August, 2022;
originally announced August 2022.
-
Analogue-digital systems and the modular decomposition of physical behaviour
Authors:
E. J. Beggs,
J. V. Tucker
Abstract:
We take a fresh look at analogue-digital systems focussing on their physical behaviour. We model a general analogue-digital system as a physical process controlled by an algorithm by viewing the physical process as physical oracle to the algorithm, generalising the notion of Turing. We develop a theoretical framework for the specification and analysis of such systems that combines five semantical…
▽ More
We take a fresh look at analogue-digital systems focussing on their physical behaviour. We model a general analogue-digital system as a physical process controlled by an algorithm by viewing the physical process as physical oracle to the algorithm, generalising the notion of Turing. We develop a theoretical framework for the specification and analysis of such systems that combines five semantical notions: actual physical behaviour, measured behaviour, predicted behaviour, computed behaviour and exceptional behaviour. Next, we consider the more general and applicable situation of complex processes that exhibit several distinct modes of physical behaviour. Thus, for 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 with its own physical oracle. The models may be of disparate kinds and, furthermore, not all physical modes may even have a reliable model. We address the questions: How do we specify algorithms and software that monitor or govern a complex physical situation with many physical modes? How do we specify a portfolio of modes, and the computational problem of transitioning from using one mode to another mode as physical behaviour changes? We propose a general definition of an analogue-digital system with modes, and 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 data type and mode selection functions. We illustrate the ideas of physical modes and our theory by reflecting on simple examples, including driverless racing cars.
△ Less
Submitted 16 April, 2018;
originally announced April 2018.
-
Multivariate information measures: an experimentalist's perspective
Authors:
Nicholas Timme,
Wesley Alford,
Benjamin Flecker,
John M. Beggs
Abstract:
Information theory is widely accepted as a powerful tool for analyzing complex systems and it has been applied in many disciplines. Recently, some central components of information theory - multivariate information measures - have found expanded use in the study of several phenomena. These information measures differ in subtle yet significant ways. Here, we will review the information theory behin…
▽ More
Information theory is widely accepted as a powerful tool for analyzing complex systems and it has been applied in many disciplines. Recently, some central components of information theory - multivariate information measures - have found expanded use in the study of several phenomena. These information measures differ in subtle yet significant ways. Here, we will review the information theory behind each measure, as well as examine the differences between these measures by applying them to several simple model systems. In addition to these systems, we will illustrate the usefulness of the information measures by analyzing neural spiking data from a dissociated culture through early stages of its development. We hope that this work will aid other researchers as they seek the best multivariate information measure for their specific research goals and system. Finally, we have made software available online which allows the user to calculate all of the information measures discussed within this paper.
△ Less
Submitted 29 August, 2012; v1 submitted 28 November, 2011;
originally announced November 2011.
-
Limits to measurement in experiments governed by algorithms
Authors:
E. J. Beggs,
J. F. Costa,
J. V. Tucker
Abstract:
We pose the following question: If a physical experiment were to be completely controlled by an algorithm, what effect would the algorithm have on the physical measurements made possible by the experiment?
In a programme to study the nature of computation possible by physical systems, and by algorithms coupled with physical systems, we have begun to analyse (i) the algorithmic nature of experi…
▽ More
We pose the following question: If a physical experiment were to be completely controlled by an algorithm, what effect would the algorithm have on the physical measurements made possible by the experiment?
In a programme to study the nature of computation possible by physical systems, and by algorithms coupled with physical systems, we have begun to analyse (i) the algorithmic nature of experimental procedures, and (ii) the idea of using a physical experiment as an oracle to Turing Machines. To answer the question, we will extend our theory of experimental oracles in order to use Turing machines to model the experimental procedures that govern the conduct of physical experiments. First, we specify an experiment that measures mass via collisions in Newtonian Dynamics; we examine its properties in preparation for its use as an oracle. We start to classify the computational power of polynomial time Turing machines with this experimental oracle using non-uniform complexity classes. Second, we show that modelling an experimenter and experimental procedure algorithmically imposes a limit on what can be measured with equipment. Indeed, the theorems suggest a new form of uncertainty principle for our knowledge of physical quantities measured in simple physical experiments. We argue that the results established here are representative of a huge class of experiments.
△ Less
Submitted 19 November, 2009;
originally announced November 2009.
