Leadership is not really about authority and asserting hierarchical command and control, but about inspiring and enabling people to work towards a shared purpose or goal. Leadership is about providing a coherent purpose for the organisation. Traditional management focuses on compliance, control, and accountability, often imposed from above, whereas cybernetic governance is about self-regulation coherent with a purpose, and adaptation to a changing environment. Unlike Artificial Intelligence that struggles with transparency, our approach is to use simple tools to capture these purposes, keeping them front and centre of change management and Business Process Design. Our approach is unashamedly cybernetic in outlook, taking a systems-based approach to modelling with a focus on identifying opportunities for feedback control.

We present a novel form of Liquid Automata, using this to simulate autopoiesis, whereby living machines self-organise in the physical realm. This simulation is based on an earlier Cellular Automaton described by Francisco Varela. The basis of Liquid Automata is a particle simulation with additional rules about how particles are transformed on collision with other particles. Unlike cellular automata, there is no fixed grid or time-step, only particles moving about and colliding with each other in a continuous space/time.

In this study, we investigate the process of generating single-sentence representations for the purpose of Dialogue Act (DA) classification, including several aspects of text pre-processing and input representation which are often overlooked or underreported within the literature, for example, the number of words to keep in the vocabulary or input sequences. We assess each of these with respect to two DA-labelled corpora, using a range of supervised models, which represent those most frequently applied to the task. Additionally, we compare context-free word embedding models with that of transfer learning via pre-trained language models, including several based on the transformer architecture, such as Bidirectional Encoder Representations from Transformers (BERT) and XLNET, which have thus far not been widely explored for the DA classification task. Our findings indicate that these text pre-processing considerations do have a statistically significant effect on classification accuracy. Notably, we found that viable input sequence lengths, and vocabulary sizes, can be much smaller than is typically used in DA classification experiments, yielding no significant improvements beyond certain thresholds. We also show that in some cases the contextual sentence representations generated by language models do not reliably outperform supervised methods. Though BERT, and its derivative models, do represent a significant improvement over supervised approaches, and much of the previous work on DA classification.

ABSTRACT We present the Conversation Analysis Modeling Schema (CAMS), a novel dialogue labeling schema that combines the Conversation Analysis concept of Adjacency Pairs, with Dialogue Acts. The aim is to capture both the semantic and syntactic structure of dialogue, in a format that is independent of the domain or topic, and which facilitates the computational modeling of dialogue. A labeling task undertaken by novice annotators is used to evaluate its efficacy on a selection of task-oriented and non-task-oriented dialogs, and to measure inter-annotator agreement. To deepen the “human-factors” analysis we also record and examine users’ self-reported confidence scores and average utterance annotation times. Inter-annotator agreement is shown to be higher for task-oriented dialogs than non-task-oriented, though the structure of the dialogue itself has a more significant impact. We further examine the assumptions around expected agreement for two weighted agreement coefficients, Alpha and Beta, and show that annotators assign labels using similar probability distributions, small variations can result in large differences in agreement values between biased and unbiased measures.

IT helpdesks are charged with the task of responding quickly to user queries. To give the user confidence that their query matters, the helpdesk will auto-reply to the user with confirmation that their query has been received and logged. This auto-reply may include generic ‘boiler-plate’ text that addresses common problems of the day, with relevant information and links. The approach explored here is to tailor the content of the auto-reply to the user’s problem, so as to increase the relevance of the information included. Problem classification is achieved by training a neural network on a suitable corpus of IT helpdesk email data. While this is no substitute for follow-up by helpdesk agents, the aim is that this system will provide a practical stop-gap.

Task-oriented dialogue systems (TODS) are continuing to rise in popularity as various industries find ways to effectively harness their capabilities, saving both time and money. However, even state-of-the-art TODS are not yet reaching their full potential. TODS typically have a primary design focus on completing the task at hand, so the metric of task-resolution should take priority. Other conversational quality attributes that may point to the success, or otherwise, of the dialogue, may be ignored. This can cause interactions between human and dialogue system that leave the user dissatisfied or frustrated. This paper explores the literature on evaluative frameworks of dialogue systems and the role of conversational quality attributes in dialogue systems, looking at if, how, and where they are utilised, and examining their correlation with the performance of the dialogue system.

The principle of biological autonomy, introduced by Francisco J. Varela, addresses the dilemma of Cartesian mind-body dualism by re-casting mind and body, or subject and object, observer and observed, not as irreconcilable categories, but as complementary perspectives on the same biological phenomena. Indeed, this distinction between self and non-self may be seen as a necessary pre-condition for autonomy. An autonomous system is self-governing in that it is concerned with preserving its unique character, or unity. Furthermore, an autonomous system is operationally closed in that it forms a self-referential network without reference to an external world. This paper develops these ideas in relation to thinking about embodied, enactive robotics. As well as being constructed artefacts, what is it to look at robots as truly autonomous agents? In this context we begin to explore the concept of operational closure analytically. We utilise natural connectivity as a quantitative measure of the cyclicity of these operationally closed internal processes. In doing so we discover that increased natural connectivity of an autonomous system confers a greater behavioural robustness when it is coupled with the external world.

AbstractBackgroundPeople in the later stages of dementia often express their needs and distress through their behaviour in ways that challenge care staff to respond appropriately. Technological solutions may help staff to recognise this distress at an early stage and take preventative action. The aim of this project is to develop and test the feasibility of using wearable technologies and machine learning algorithms to identify stress and agitation.MethodTwo laboratory‐based feasibility studies were undertaken with healthy adult volunteers. In study 1, 20 participants took part in a computer‐based exercise designed to elicit stress. In study 2, 20 participants completed eight different physical activities which a person living with dementia might undertake during a typical day, plus two additional activities which were designed to be positively‐ and negatively‐stimulating. In both studies, participants’ physiological responses were recorded using wearables – in study 1, using commercially‐available devices worn around the wrist and waist, and in study 2, using an additional device developed by the researcher, worn in a sock. Participants also provided information on their affective states – in study 1 they rated their stress, and in study 2 they completed the Positive and Negative Affect Schedule (PANAS) before and after each activity. Participants’ responses were used to train and evaluate different binary classifiers for their ability to recognise reported stress in the physiological data.ResultsIn study 1, the K‐nearest neighbours algorithm and an ensemble classifier called bagged trees recognised periods when participants reported feeling most stressed with accuracy rates exceeding 90% (98% sensitivity, >99% specificity), using five‐fold validation. However, in study 2 which used a different group of participants, the algorithms trained in study 1 failed to reliably identify the negatively‐stimulating activity from the other activities.ConclusionTechnological solutions can facilitate a proactive approach to the management of stress and agitation in dementia. However, for these kinds of solutions to be effective in practice, they must be able to generalise to new users and different environments including uncontrolled environments. These kinds of solutions also raise ethical considerations, such as privacy and data security, which are particularly important when users may lack capacity.

The behaviour of a reactive, line-following robot is analysed using modal logic. This provides an approach that is complementary to numerical simulation, allowing us to explore the qualitative state-space of the robot coupled with its environment. The envisionment of this state-space can be described as a Kripke model, and model-checking tools enable us to analyse this model to search for stable equilibria that contain goal states.

The identification of Dialogue Act’s (DA) is an important aspect in determining the meaning of an utterance for many applications that require natural language understanding, and recent work using recurrent neural networks (RNN) has shown promising results when applied to the DA classification problem. This work presents a novel probabilistic method of utterance representation and describes a RNN sentence model for out-of-context DA Classification. The utterance representations are generated from keywords selected for their frequency association with certain DA’s. The proposed probabilistic representations are applied to the Switchboard DA corpus and performance is compared with pre-trained word embeddings using the same baseline RNN model. The results indicate that the probabilistic method achieves 75.48% overall accuracy and an improvement over the word embedding representations of 1.8%. This demonstrates the potential utility of using statistical utterance representations, that are able to capture word-DA relationships, for the purpose of DA classification.

The Bristol LoRaWAN Network is a low power radio network for the Internet of Things, based on LoRaWAN and utilising The Things Network. LoRaWAN is a Low Power Wide Area Network (LPWAN) specification intended for wireless battery operated Things in regional, national, or global networks. LoRaWAN targets key requirements of internet of things such as secure bi-directional communication, mobility and localisation services. A 2015 pilot programme in the city of Amsterdam aimed to cover the entire city with just 10 wireless gateways. LoRaWAN Bristol aims to replicate this experiment in the UK.

Using Punch and Judy as a story domain, we describe an interactive puppet show, where the flow and content of the story can be influenced by the actions of the audience. As the puppet show is acted out, the audience reacts to events by cheering or booing the characters. This affects the agents’ emotional state, potentially causing them to change their actions, altering the course of the narrative. An institutional normative model is used to constrain the narrative so that it remains consistent with the Punch and Judy canon. Through this vignette of a socio-technical system (STS), comprising human and software actors, an institutional model – derived from narrative theory – and (simplistic) technological interaction artifacts, we begin to be able to explore some of the issues that can arise in STS through the prism of the World-Institution-Technology (WIT) model.

This paper develops models for single recurrent neural circuits, known as autapses, where the axon of a neuron forms synapses onto its own dendrites. Once thought to be curiosities or artefacts of growing cells in vitro, autapses play a key role in the operation of Central Pattern Generators and the cortex where they may function as a simple short-term memory. Biologically plausible, idealized models of the autapse are able to produce arrhythmic, sustained behaviours in ‘neural vehicles’. Detailed models are developed to show how excitatory autapses may support both bistability and monostability.

In Design for a Brain, W. Ross Ashby speculates about the possibility of creating a mobile homeostat "with its critical states set so that it seeks situations of high illumination." This paper explores a realization of Ashby's homeostat within a simulated robot and environment exploring the question as to whether the classic homeostat architecture is able to adapt to this environment. Remaining faithful to the physical design of Ashby's device, this simulation enables us to quantitatively evaluate Ashby's proposition that homeostasis can be achieved through ultrastability. Following his law of requisite variety it is demonstrated that increasing the number of units increases the time taken to reach equilibrium, and that conversely, reducing internal connectivity reduces the time taken to reach equilibrium.

A narrative world can be viewed as a form of society in which characters follow a set of social norms whose collective function is to guide the characters through (the creation of) a story arc and reach some conclusion. By modelling the rules of a narrative using norms, we can govern the actions of agents that act out the characters in a story. Agents are given sets of permitted actions and obligations to fulfil based on their and the story's current situation. However, a way to describe stories in terms of social norms is needed. Existing formalisms for narrative do not work at multiple layers of abstraction, and do not provide a rich enough vocabulary for describing stories. We use story tropes as a means of building reusable story components with which we describe the social norms that govern our storyworld agents.

Authoring a consistent interactive narrative is difficult without exhaustively specifying all possible deviations from the main path of a story. When automatically generating new story paths, it is important to be able to check these paths for consistency with the narrative world. We present a method of describing the structure of a story as a Kripke structure using Interval Temporal Logic. This allows the model checking of each possible telling of the narrative for consistency with the story world, as well as the ability to construct re-usable story components at different levels of abstraction. This is the first step towards building a fully checkable framework for building story components using modal logic.