A pseudo-marginal sequential Monte Carlo algorithm for random effects models in Bayesian sequential design

A computationally efficient sequential Monte Carlo algorithm is proposed for the sequential design of experiments for the collection of block data described by mixed effects models. The difficulty in applying a sequential Monte Carlo algorithm in such settings is the need to evaluate the observed data likelihood, which is typically intractable for all but linear Gaussian models. To overcome this difficulty, we propose to unbiasedly estimate the likelihood, and perform inference and make decisions based on an exact-approximate algorithm. Two estimates are proposed: using Quasi Monte Carlo methods and using the Laplace approximation with importance sampling. Both of these approaches can be computationally expensive, so we propose exploiting parallel computational architectures to ensure designs can be derived in a timely manner. We also extend our approach to allow for model uncertainty. This research is motivated by important pharmacological studies related to the treatment of critically ill patients.

Homing by path integration when a locomotion trajectory crosses itself

Path integration is a process with which navigators derive their current position and orientation by integrating self-motion signals along a locomotion trajectory. It has been suggested that path integration becomes disproportionately erroneous when the trajectory crosses itself. However, there is a possibility that this previous finding was confounded by effects of the length of a traveled path and the amount of turns experienced along the path, two factors that are known to affect path integration performance. The present study was designed to investigate whether the crossover of a locomotion trajectory truly increases errors of path integration. In an experiment, blindfolded human navigators were guided along four paths that varied in their lengths and turns, and attempted to walk directly back to the beginning of the paths. Only one of the four paths contained a crossover. Results showed that errors yielded from the path containing the crossover were not always larger than those observed in other paths, and the errors were attributed solely to the effects of longer path lengths or greater degrees of turns. These results demonstrated that path crossover does not always cause significant disruption in path integration processes. Implications of the present findings for models of path integration are discussed.

Incorporating human-factors in car-following models : a review of recent developments and research needs

Over the past decades there has been a considerable development in the modeling of car-following (CF) behavior as a result of research undertaken by both traffic engineers and traffic psychologists. While traffic engineers seek to understand the behavior of a traffic stream, traffic psychologists seek to describe the human abilities and errors involved in the driving process. This paper provides a comprehensive review of these two research streams. It is necessary to consider human-factors in {CF} modeling for a more realistic representation of {CF} behavior in complex driving situations (for example, in traffic breakdowns, crash-prone situations, and adverse weather conditions) to improve traffic safety and to better understand widely-reported puzzling traffic flow phenomena, such as capacity drop, stop-and-go oscillations, and traffic hysteresis. While there are some excellent reviews of {CF} models available in the literature, none of these specifically focuses on the human factors in these models. This paper addresses this gap by reviewing the available literature with a specific focus on the latest advances in car-following models from both the engineering and human behavior points of view. In so doing, it analyses the benefits and limitations of various models and highlights future research needs in the area.

Theorizing the use of multiple conceptual models in combination

Conceptual modelling continues to be an important means for graphically capturing the requirements of an information system. Observations of modelling practice suggest that modellers often use multiple conceptual models in combination, because they articulate different aspects of real-world domains. Yet, the available empirical as well as theoretical research in this area has largely studied the use of single models, or single modelling grammars. We develop a Theory of Combined Ontological Coverage by extending an existing theory of ontological expressiveness of conceptual modelling grammars. Our new theory posits that multiple conceptual models are used to increase the maximum coverage of the real-world domain being modelled, whilst trying to minimize the ontological overlap between the models. We illustrate how the theory can be applied to analyse sets of conceptual models. We develop three propositions of the theory about evaluations of model combinations in terms of users’ selection, understandability and usefulness of conceptual models.

MIMO and SISO identification of radiation force terms for models of marine structures in waves

This paper presents a discussion on the use of MIMO and SISO techniques for identification of the radiation force terms in models for surface vessels. We compare and discuss two techniques recently proposed in literature for this application: time domain identification and frequency domain identification. We compare the methods in terms of estimates model order, accuracy of the fit, use of the available information, and ease of use and implementation.

Functional magnetic resonance brain imaging of executive cognitive performance in young first-episode schizophrenia patients and age-matched long-term cannabis users

Converging evidence from epidemiological, clinical and neuropsychological research suggests a link between cannabis use and increased risk of psychosis. Long-term cannabis use has also been related to deficit-like “negative” symptoms and cognitive impairment that resemble some of the clinical and cognitive features of schizophrenia. The current functional brain imaging study investigated the impact of a history of heavy cannabis use on impaired executive function in first-episode schizophrenia patients. Whilst performing the Tower of London task in a magnetic resonance imaging scanner, event-related blood oxygenation level-dependent (BOLD) brain activation was compared between four age and gender-matched groups: 12 first-episode schizophrenia patients; 17 long-term cannabis users; seven cannabis using first-episode schizophrenia patients; and 17 healthy control subjects. BOLD activation was assessed as a function of increasing task difficulty within and between groups as well as the main effects of cannabis use and the diagnosis of schizophrenia. Cannabis users and non-drug using first-episode schizophrenia patients exhibited equivalently reduced dorsolateral prefrontal activation in response to task difficulty. A trend towards additional prefrontal and left superior parietal cortical activation deficits was observed in cannabis-using first-episode schizophrenia patients while a history of cannabis use accounted for increased activation in the visual cortex. Cannabis users and schizophrenia patients fail to adequately activate the dorsolateral prefrontal cortex, thus pointing to a common working memory impairment which is particularly evident in cannabis-using first-episode schizophrenia patients. A history of heavy cannabis use, on the other hand, accounted for increased primary visual processing, suggesting compensatory imagery processing of the task.

Design after design to bridge between people living with cognitive or sensory impairments, their friends and proxies

A carer or teacher often plays the role of proxy or spokesperson for a person living with an intellectual disability or form of cognitive or sensory impairment. Our research undertook co-design with people living with cognitive and sensory impairments and their proxies in order to explore new ways of facilitating communication. We developed simple functioning interactive prototypes to support people with a diverse range of competencies to communicate and explore their use. Deployment of the prototypes enabled use, appropriation and design after design by our two participant groups; adults living with cognitive or sensory impairments and children identified with language delays and autism spectrum disorder. The prototypes supported concrete expression of likes, dislikes, capabilities, emotional wants and needs and forms of expression that hitherto had not been fostered, further informing design. Carers and designers were surprised at the ways in which the technology was used and how it fostered new forms of social interaction and expression. We elaborate on how design after design can be an effective approach for engaging people living with intellectual disabilities, giving them greater capacity for expression and power in design and offering the potential to expand and deepen their social relationships.

Principles of goal-directed spatial robot navigation in biomimetic models

Mobile robots and animals alike must effectively navigate their environments in order to achieve their goals. For animals goal-directed navigation facilitates finding food, seeking shelter or migration; similarly robots perform goal-directed navigation to find a charging station, get out of the rain or guide a person to a destination. This similarity in tasks extends to the environment as well; increasingly, mobile robots are operating in the same underwater, ground and aerial environments that animals do. Yet despite these similarities, goal-directed navigation research in robotics and biology has proceeded largely in parallel, linked only by a small amount of interdisciplinary research spanning both areas. Most state-of-the-art robotic navigation systems employ a range of sensors, world representations and navigation algorithms that seem far removed from what we know of how animals navigate; their navigation systems are shaped by key principles of navigation in ‘real-world’ environments including dealing with uncertainty in sensing, landmark observation and world modelling. By contrast, biomimetic animal navigation models produce plausible animal navigation behaviour in a range of laboratory experimental navigation paradigms, typically without addressing many of these robotic navigation principles. In this paper, we attempt to link robotics and biology by reviewing the current state of the art in conventional and biomimetic goal-directed navigation models, focusing on the key principles of goal-oriented robotic navigation and the extent to which these principles have been adapted by biomimetic navigation models and why.

Computation of Mathematical Models for Complex Industrial Processes

Designed for undergraduate and postgraduate students, academic researchers and industrial practitioners, this book provides comprehensive case studies on numerical computing of industrial processes and step-by-step procedures for conducting industrial computing. It assumes minimal knowledge in numerical computing and computer programming, making it easy to read, understand and follow. Topics discussed include fundamentals of industrial computing, finite difference methods, the Wavelet-Collocation Method, the Wavelet-Galerkin Method, High Resolution Methods, and comparative studies of various methods. These are discussed using examples of carefully selected models from real processes of industrial significance. The step-by-step procedures in all these case studies can be easily applied to other industrial processes without a need for major changes and thus provide readers with useful frameworks for the applications of engineering computing in fundamental research problems and practical development scenarios.

Fourier active appearance models

Gaining invariance to camera and illumination variations has been a well investigated topic in Active Appearance Model (AAM) fitting literature. The major problem lies in the inability of the appearance parameters of the AAM to generalize to unseen conditions. An attractive approach for gaining invariance is to fit an AAM to a multiple filter response (e.g. Gabor) representation of the input image. Naively applying this concept with a traditional AAM is computationally prohibitive, especially as the number of filter responses increase. In this paper, we present a computationally efficient AAM fitting algorithm based on the Lucas-Kanade (LK) algorithm posed in the Fourier domain that affords invariance to both expression and illumination. We refer to this as a Fourier AAM (FAAM), and show that this method gives substantial improvement in person specific AAM fitting performance over traditional AAM fitting methods.