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.

Peripheral, central and coercive routes for promoting Enterprise Social Networks

Organisations employ Enterprise Social Networks (ESNs) (such as Yammer) expecting better intra-organisational communication and collaboration. However, ESNs are struggling to gain momentum and wide adoption among users. Promoting user participation is a challenge, particularly in relation to lurkers – the silent ESN members who do not contribute any content. Building on behaviour change research, we propose a three-route model consisting of the central, peripheral and coercive routes of influence that depict users’ cognitive strategies, and we examine how management interventions (e.g. sending promotional emails) impact users’ beliefs and (consequent) posting and lurking behaviours in ESNs. Furthermore, we identify users’ salient motivations to lurk or post. We employ a multi-method research design to conceptualise, operationalise and validate the research model. This study has implications for academics and practitioners regarding the nature, patterns and outcomes of management interventions in prompting ESN.

Informal caring networks for people at end of life: Building social capital in Australian communities

The care of a person living at home near the end of their life is predominantly provided by family carers with the support of health services such as palliative care. In addition, informal caring networks also contribute at times to the support to the dying person and their carer. In this way, these networks can promote social capital in the communities from which they are drawn. This social approach to end of life care enhances community capacity to provide support to those dying at home and their carers. This article examines relevant published literature to explore the conceptual foundations of informal caring networks, examining the place of social capital and community development in the provision of end of life care at home, particularly in the Australian context.

Integrating PV systems into distribution networks with battery energy storage systems

Integrating Photovoltaic (PV) systems with battery energy storage in the distribution network will be essential to allow for continued uptake of domestic PV system installations. With increasing concerns regarding environmental and climate change issues, incorporating sources of renewable energy into power networks across the world will be key for a sustainable future. Australia is well placed to utilise solar energy as a significant component of its future energy generation and within the last 5 years there has been a rapid growth in the penetration levels seen by the grid. This growth of PV systems is causing a number of issues including intermittency of supply, negative power flow and voltage rises. Using the simulator tool GridLAB-D with a model of a typical South-East Queensland (SEQ) 11 kV distribution feeder, the effect of various configurations of PV systems have been offset with Battery Energy Storage Systems (BESS). From this, combinations of PV and storage that are most effective at mitigating the issues were explored.

Do good things and talk about them : a theory of academics usage of enterprise social networks for impression management tactics

Enterprise social networks provide benefits especially for knowledge-intensive work as they enable communication, collaboration and knowledge exchange. These platforms should therefore lead to increased adoption and use by knowledge-intensive workers such as consultants or indeed researchers. Our interest is in ascertaining whether scientific researchers use enterprise social networks as part of their work practices. This focus is motivated by an apparent schism between a need for researchers to exchange knowledge and profile themselves, and the aversion to sharing breakthrough ideas and joining in an ever-increasing publishing and marketing game. We draw on research on academic work practices and impression management to develop a model of academics’ ESN usage for impression management tactics. We describe important constructs of our model, offer strategies for their operationalization and give an outlook to our ongoing empirical study of the use of an ESN platform by 20 schools across six faculties at an Australian university.

Anomaly detection in online social networks : using data-mining techniques and fuzzy logic

This research is a step forward in improving the accuracy of detecting anomaly in a data graph representing connectivity between people in an online social network. The proposed hybrid methods are based on fuzzy machine learning techniques utilising different types of structural input features. The methods are presented within a multi-layered framework which provides the full requirements needed for finding anomalies in data graphs generated from online social networks, including data modelling and analysis, labelling, and evaluation.

Understanding Uncertainties in Non-Linear Population Trajectories: A Bayesian Semi-Parametric Hierarchical Approach to Large-Scale Surveys of Coral Cover

Recently, attempts to improve decision making in species management have focussed on uncertainties associated with modelling temporal fluctuations in populations. Reducing model uncertainty is challenging; while larger samples improve estimation of species trajectories and reduce statistical errors, they typically amplify variability in observed trajectories. In particular, traditional modelling approaches aimed at estimating population trajectories usually do not account well for nonlinearities and uncertainties associated with multi-scale observations characteristic of large spatio-temporal surveys. We present a Bayesian semi-parametric hierarchical model for simultaneously quantifying uncertainties associated with model structure and parameters, and scale-specific variability over time. We estimate uncertainty across a four-tiered spatial hierarchy of coral cover from the Great Barrier Reef. Coral variability is well described; however, our results show that, in the absence of additional model specifications, conclusions regarding coral trajectories become highly uncertain when considering multiple reefs, suggesting that management should focus more at the scale of individual reefs. The approach presented facilitates the description and estimation of population trajectories and associated uncertainties when variability cannot be attributed to specific causes and origins. We argue that our model can unlock value contained in large-scale datasets, provide guidance for understanding sources of uncertainty, and support better informed decision making

Recent advances in modeling the vulnerability of transportation networks

It is well known that, for major infrastructure networks such as electricity, gas, railway, road, and urban water networks, disruptions at one point have a knock on effect throughout the network. There is an impressive amount of individual research projects examining the vulnerability of critical infrastructure network. However, there is little understanding of the totality of the contribution made by these projects and their interrelationships. This makes their review a difficult process for both new and existing researchers in the field. To address this issue, a two-step literature review process is used, to provide an overview of the vulnerability of the transportation network in terms of four main themes - research objective, transportation mode, disruption scenario and vulnerability indicator –involving the analysis of related articles from 2001 to 2013. Two limitations of existing research are identified: (1) the limited amount of studies relating to multi-layer transportation network vulnerability analysis, and (2) the lack of evaluation methods to explore the relationship between structure vulnerability and dynamical functional vulnerability. In addition to indicating that more attention needs to be paid to these two aspects in future, the analysis provides a new avenue for the discovery of knowledge, as well as an improved understanding of transportation network vulnerability.

Multi-agent systems for modeling high penetration photovoltaic system impacts in distribution networks

This paper reviews the use of multi-agent systems to model the impacts of high levels of photovoltaic (PV) system penetration in distribution networks and presents some preliminary data obtained from the Perth Solar City high penetration PV trial. The Perth Solar City trial consists of a low voltage distribution feeder supplying 75 customers where 29 consumers have roof top photovoltaic systems. Data is collected from smart meters at each consumer premises, from data loggers at the transformer low voltage (LV) side and from a nearby distribution network SCADA measurement point on the high voltage side (HV) side of the transformer. The data will be used to progressively develop MAS models.

The new supply chains : the Multichannel Networks

Digital innovation is transforming the media and entertainment industries. The professionalization of YouTube’s platform is paradigmatic of that change. The 100 original channel initiative launched in late 2011 was designed to transform YouTube’s brand through production of a high volume of quality premium video content that would more deeply engage its audience base and in the process attract big advertisers. An unanticipated by-product has been the rapid growth of a wave of aspiring next-generation digital media companies from within the YouTube ecosystem. Fuelled by early venture capital some have ambitious goals to become global media corporations in the online video space. A number of larger MCNs (Multi-Channel Networks) - BigFrame, Machinima, Fullscreen, AwesomenessTV, Maker Studios , Revision3 and DanceOn - have attracted interest from media incumbents like Warner Brothers, DreamWorks, Discovery, Bertlesmann, Comcast and AMC, and two larger MCNs Alloy and Break Media have merged. This indicates that a shakeout is underway in these new online supply chains, after rapid initial growth. The higher profile MCNs seek to rapidly develop scale economies in online distribution and facilitate audience growth for their member channels, helping channels optimize monetization, develop sustainable business models and to facilitate producer-collaboration within a growing online community of like-minded content creators. Some MCNs already attract far larger online audiences than any national TV network. The speed with which these developments have occurred is reminiscent of the 1910s, when Hollywood studios first emerged and within only a few years replaced the incumbent film studios as the dominant force within the film industry.

Approximation of Bayesian predictive p-values with regression ABC

In the Bayesian framework a standard approach to model criticism is to compare some function of the observed data to a reference predictive distribution. The result of the comparison can be summarized in the form of a p-value, and it's well known that computation of some kinds of Bayesian predictive p-values can be challenging. The use of regression adjustment approximate Bayesian computation (ABC) methods is explored for this task. Two problems are considered. The first is the calibration of posterior predictive p-values so that they are uniformly distributed under some reference distribution for the data. Computation is difficult because the calibration process requires repeated approximation of the posterior for different data sets under the reference distribution. The second problem considered is approximation of distributions of prior predictive p-values for the purpose of choosing weakly informative priors in the case where the model checking statistic is expensive to compute. Here the computation is difficult because of the need to repeatedly sample from a prior predictive distribution for different values of a prior hyperparameter. In both these problems we argue that high accuracy in the computations is not required, which makes fast approximations such as regression adjustment ABC very useful. We illustrate our methods with several samples.

Contributions to Bayesian experimental design

This thesis progresses Bayesian experimental design by developing novel methodologies and extensions to existing algorithms. Through these advancements, this thesis provides solutions to several important and complex experimental design problems, many of which have applications in biology and medicine. This thesis consists of a series of published and submitted papers. In the first paper, we provide a comprehensive literature review on Bayesian design. In the second paper, we discuss methods which may be used to solve design problems in which one is interested in finding a large number of (near) optimal design points. The third paper presents methods for finding fully Bayesian experimental designs for nonlinear mixed effects models, and the fourth paper investigates methods to rapidly approximate the posterior distribution for use in Bayesian utility functions.

Pre-processing for approximate Bayesian computation in image analysis

Most of the existing algorithms for approximate Bayesian computation (ABC) assume that it is feasible to simulate pseudo-data from the model at each iteration. However, the computational cost of these simulations can be prohibitive for high dimensional data. An important example is the Potts model, which is commonly used in image analysis. Images encountered in real world applications can have millions of pixels, therefore scalability is a major concern. We apply ABC with a synthetic likelihood to the hidden Potts model with additive Gaussian noise. Using a pre-processing step, we fit a binding function to model the relationship between the model parameters and the synthetic likelihood parameters. Our numerical experiments demonstrate that the precomputed binding function dramatically improves the scalability of ABC, reducing the average runtime required for model fitting from 71 hours to only 7 minutes. We also illustrate the method by estimating the smoothing parameter for remotely sensed satellite imagery. Without precomputation, Bayesian inference is impractical for datasets of that scale.

Classification of echolocation calls from 14 species of bat by Support Vector Machines and Ensembles of Neural Networks

Calls from 14 species of bat were classified to genus and species using discriminant function analysis (DFA), support vector machines (SVM) and ensembles of neural networks (ENN). Both SVMs and ENNs outperformed DFA for every species while ENNs (mean identification rate – 97%) consistently outperformed SVMs (mean identification rate – 87%). Correct classification rates produced by the ENNs varied from 91% to 100%; calls from six species were correctly identified with 100% accuracy. Calls from the five species of Myotis, a genus whose species are considered difficult to distinguish acoustically, had correct identification rates that varied from 91 – 100%. Five parameters were most important for classifying calls correctly while seven others contributed little to classification performance.

Human vs. machine : identification of bat species from their echolocation calls by humans and by artificial neural networks

Automated remote ultrasound detectors allow large amounts of data on bat presence and activity to be collected. Processing of such data involves identifying bat species from their echolocation calls. Automated species identification has the potential to provide more consistent, predictable, and potentially higher levels of accuracy than identification by humans. In contrast, identification by humans permits flexibility and intelligence in identification, as well as the incorporation of features and patterns that may be difficult to quantify. We compared humans with artificial neural networks (ANNs) in their ability to classify short recordings of bat echolocation calls of variable signal to noise ratios; these sequences are typical of those obtained from remote automated recording systems that are often used in large-scale ecological studies. We presented 45 recordings (1–4 calls) produced by known species of bats to ANNs and to 26 human participants with 1 month to 23 years of experience in acoustic identification of bats. Humans correctly classified 86% of recordings to genus and 56% to species; ANNs correctly identified 92% and 62%, respectively. There was no significant difference between the performance of ANNs and that of humans, but ANNs performed better than about 75% of humans. There was little relationship between the experience of the human participants and their classification rate. However, humans with <1 year of experience performed worse than others. Currently, identification of bat echolocation calls by humans is suitable for ecological research, after careful consideration of biases. However, improvements to ANNs and the data that they are trained on may in future increase their performance to beyond those demonstrated by humans.