Prediction with limited advice and multiarmed bandits with paid observations

We study two problems of online learning under restricted information access. In the first problem, prediction with limited advice, we consider a game of prediction with expert advice, where on each round of the game we query the advice of a subset of M out of N experts. We present an algorithm that achieves O(√(N/M)TlnN ) regret on T rounds of this game. The second problem, the multiarmed bandit with paid observations, is a variant of the adversarial N-armed bandit game, where on round t of the game we can observe the reward of any number of arms, but each observation has a cost c. We present an algorithm that achieves O((cNlnN) 1/3 T2/3+√TlnN ) regret on T rounds of this game in the worst case. Furthermore, we present a number of refinements that treat arm- and time-dependent observation costs and achieve lower regret under benign conditions. We present lower bounds that show that, apart from the logarithmic factors, the worst-case regret bounds cannot be improved.

Structured contract strategies for capital and operations expenditure projects in the oil and gas industry

This project was a step forward in developing a 'descriptive theory' of contracting in the oil and gas industry that reflects the operating environment in which the project manager operates. This study investigates the existing processes and methods used in establishing contracts which are very often prescriptive, and not always appropriate or optimal for a given situation. This study contributes to contracting effectiveness or optimal contracting in the oil and gas industry.

Comparing different approach and avoidance models of learning and personality in the prediction of work, university, and leadership outcomes

Jackson (2005) developed a hybrid model of personality and learning, known as the learning styles profiler (LSP) which was designed to span biological, socio-cognitive, and experiential research foci of personality and learning research. The hybrid model argues that functional and dysfunctional learning outcomes can be best understood in terms of how cognitions and experiences control, discipline, and re-express the biologically based scale of sensation-seeking. In two studies with part-time workers undertaking tertiary education (N equals 137 and 58), established models of approach and avoidance from each of the three different research foci were compared with Jackson's hybrid model in their predictiveness of leadership, work, and university outcomes using self-report and supervisor ratings. Results showed that the hybrid model was generally optimal and, as hypothesized, that goal orientation was a mediator of sensation-seeking on outcomes (work performance, university performance, leader behaviours, and counterproductive work behaviour). Our studies suggest that the hybrid model has considerable promise as a predictor of work and educational outcomes as well as dysfunctional outcomes.

Traffic incident clearance time and arrival time prediction based on hazard models

Accurate prediction of incident duration is not only important information of Traffic Incident Management System, but also an ffective input for travel time prediction. In this paper, the hazard based prediction odels are developed for both incident clearance time and arrival time. The data are obtained from the Queensland Department of Transport and Main Roads’ STREAMS Incident Management System (SIMS) for one year ending in November 2010. The best fitting distributions are drawn for both clearance and arrival time for 3 types of incident: crash, stationary vehicle, and hazard. The results show that Gamma, Log-logistic, and Weibull are the best fit for crash, stationary vehicle, and hazard incident, respectively. The obvious impact factors are given for crash clearance time and arrival time. The quantitative influences for crash and hazard incident are presented for both clearance and arrival. The model accuracy is analyzed at the end.

Predictive validity of accelerometer prediction equations for energy expenditure (EE) during overland walking and running in children and adolescents

Purpose The aim of this study was to assess the predictive validity of three accelerometer prediction equations (Freedson et aL, 1997; Trost et aL, 1998; Puyau et al., 2002) for energy expenditure (EE) during overland walking and running in children and adolescents. Methods 45 healthy children and adolescents aged 10-18 completed the following protocol, each task 5-mins in duration, with a 5-min rest period in between; walking normally; walking briskly; running easily and running fast. During each task participants wore MTI (WAM 7164) Actigraphs on the left and right hips. VO2 was monitored breath by breath using the Cosmed K4b2 portable indirect calorimetry system. For each prediction equation, difference scores were calculated as EE measured minus EE predicted. The percentage of 1-min epochs correctly categorized as light (<3 METs), moderate (3-5.9 METs), and vigorous (≥6 METS) was also calculated. Results The Freedson and Trost equations consistently overestimated MET level. The level of overestimation was statistically significant across all tasks for the Freedson equation, and was significant for only the walking tasks for the Trost equation. The Puyau equation consistently underestimated AEE with the exception of the walking normally task. In terms of categorisation, the Freedson equation (72.8% agreement) demonstrated better agreement than the Puyau (60.6%). Conclusions These data suggest that the three accelerometer prediction equations do not accurately predict EE on a minute-by-minute basis in children and adolescents during overland walking and running. However, the cut points generated by these equations maybe useful for classifying activity as either, light, moderate, or vigorous.

Spatial prediction on a river network

This article develops methods for spatially predicting daily change of dissolved oxygen (Dochange) at both sampled locations (134 freshwater sites in 2002 and 2003) and other locations of interest throughout a river network in South East Queensland, Australia. In order to deal with the relative sparseness of the monitoring locations in comparison to the number of locations where one might want to make predictions, we make a classification of the river and stream locations. We then implement optimal spatial prediction (ordinary and constrained kriging) from geostatistics. Because of their directed-tree structure, rivers and streams offer special challenges. A complete approach to spatial prediction on a river network is given, with special attention paid to environmental exceedances. The methodology is used to produce a map of Dochange predictions for 2003. Dochange is one of the variables measured as part of the Ecosystem Health Monitoring Program conducted within the Moreton Bay Waterways and Catchments Partnership.

Conceptualising, evaluating and reporting social resilience in vulnerable regional and remote communities facing climate change in tropical Queensland : marine and tropical sciences research facility transition project final report

Regional and remote communities in tropical Queensland are among Australia’s most vulnerable in the face of climate change. At the same time, these socially and economically vulnerable regions house some of Australia’s most significant biodiversity values. Past approaches to terrestrial biodiversity management have focused on tackling biophysical interventions through the use of biophysical knowledge. An equally important focus should be placed on building regional-scale community resilience if some of the worst biodiversity impacts of climate change are to be avoided or mitigated. Despite its critical need, more systemic or holistic approaches to natural resource management have been rarely trialed and tested in a structured way. Currently, most strategic interventions in improving regional community resilience are ad hoc, not theory-based and short term. Past planning approaches have not been durable, nor have they been well informed by clear indicators. Research into indicators for community resilience has been poorly integrated within adaptive planning and management cycles. This project has aimed to resolve this problem by: * Reviewing the community and social resilience and adaptive planning literature to reconceptualise an improved framework for applying community resilience concepts; * Harvesting and extending work undertaken in MTSRF Phase 1 to identifying the learnings emerging from past MTSRF research; * Distilling these findings to identify new theoretical and practical approaches to the application of community resilience in natural resource use and management; * Reconsidering the potential interplay between a region’s biophysical and social planning processes, with a focus on exploring spatial tools to communicate climate change risk and its consequent environmental, economic and social impacts, and; * Trialling new approaches to indicator development and adaptive planning to improve community resilience, using a sub-regional pilot in the Wet Tropics. In doing so, we also looked at ways to improve the use and application of relevant spatial information. Our theoretical review drew upon the community development, psychology and emergency management literature to better frame the concept of community resilience relative to aligned concepts of social resilience, vulnerability and adaptive capacity. Firstly, we consider community resilience as a concept that can be considered at a range of scales (e.g. regional, locality, communities of interest, etc.). We also consider that overall resilience at higher scales will be influenced by resilience levels at lesser scales (inclusive of the resilience of constituent institutions, families and individuals). We illustrate that, at any scale, resilience and vulnerability are not necessarily polar opposites, and that some understanding of vulnerability is important in determining resilience. We position social resilience (a concept focused on the social characteristics of communities and individuals) as an important attribute of community resilience, but one that needs to be considered alongside economic, natural resource, capacity-based and governance attributes. The findings from the review of theory and MTSRF Phase 1 projects were synthesized and refined by the wider project team. Five predominant themes were distilled from this literature, research review and an expert analysis. They include the findings that: 1. Indicators have most value within an integrated and adaptive planning context, requiring an active co-research relationship between community resilience planners, managers and researchers if real change is to be secured; 2. Indicators of community resilience form the basis for planning for social assets and the resilience of social assets is directly related the longer term resilience of natural assets. This encourages and indeed requires the explicit development and integration of social planning within a broader natural resource planning and management framework; 3. Past indicator research and application has not provided a broad picture of the key attributes of community resilience and there have been many attempts to elicit lists of “perfect” indicators that may never be useful within the time and resource limitations of real world regional planning and management. We consider that modeling resilience for proactive planning and prediction purposes requires the consideration of simple but integrated clusters of attributes; 4. Depending on time and resources available for planning and management, the combined use of well suited indicators and/or other lesser “lines of evidence” is more flexible than the pursuit of perfect indicators, and that; 5. Index-based, collaborative and participatory approaches need to be applied to the development, refinement and reporting of indicators over longer time frames. We trialed the practical application of these concepts via the establishment of a collaborative regional alliance of planners and managers involved in the development of climate change adaptation strategies across tropical Queensland (the Gulf, Wet Tropics, Cape York and Torres Strait sub-regions). A focus on the Wet Tropics as a pilot sub-region enabled other Far North Queensland sub-region’s to participate and explore the potential extension of this approach. The pilot activities included: * Further exploring ways to innovatively communicate the region’s likely climate change scenarios and possible environmental, economic and social impacts. We particularly looked at using spatial tools to overlay climate change risks to geographic communities and social vulnerabilities within those communities; * Developing a cohesive first pass of a State of the Region-style approach to reporting community resilience, inclusive of regional economic viability, community vitality, capacitybased and governance attributes. This framework integrated a literature review, expert (academic and community) and alliance-based contributions; and * Early consideration of critical strategies that need to be included in unfolding regional planning activities with Far North Queensland. The pilot assessment finds that rural, indigenous and some urban populations in the Wet Tropics are highly vulnerable and sensitive to climate change and may require substantial support to adapt and become more resilient. This assessment finds that under current conditions (i.e. if significant adaptation actions are not taken) the Wet Tropics as a whole may be seriously impacted by the most significant features of climate change and extreme climatic events. Without early and substantive action, this could result in declining social and economic wellbeing and natural resource health. Of the four attributes we consider important to understanding community resilience, the Wet Tropics region is particularly vulnerable in two areas; specifically its economic vitality and knowledge, aspirations and capacity. The third and fourth attributes, community vitality and institutional governance are relatively resilient but are vulnerable in some key respects. In regard to all four of these attributes, however, there is some emerging capacity to manage the possible shocks that may be associated with the impacts of climate change and extreme climatic events. This capacity needs to be carefully fostered and further developed to achieve broader community resilience outcomes. There is an immediate need to build individual, household, community and sectoral resilience across all four attribute groups to enable populations and communities in the Wet Tropics region to adapt in the face of climate change. Preliminary strategies of importance to improve regional community resilience have been identified. These emerging strategies also have been integrated into the emerging Regional Development Australia Roadmap, and this will ensure that effective implementation will be progressed and coordinated. They will also inform emerging strategy development to secure implementation of the FNQ 2031 Regional Plan. Of most significance in our view, this project has taken a co-research approach from the outset with explicit and direct importance and influence within the region’s formal planning and management arrangements. As such, the research: * Now forms the foundations of the first attempt at “Social Asset” planning within the Wet Tropics Regional NRM Plan review; * Is assisting Local government at regional scale to consider aspects of climate change adaptation in emerging planning scheme/community planning processes; * Has partnered the State government (via the Department of Infrastructure and Planning and Regional Managers Coordination Network Chair) in progressing the Climate Change adaptation agenda set down within the FNQ 2031 Regional Plan; * Is informing new approaches to report on community resilience within the GBRMPA Outlook reporting framework; and * Now forms the foundation for the region’s wider climate change adaptation priorities in the Regional Roadmap developed by Regional Development Australia. Through the auspices of Regional Development Australia, the outcomes of the research will now inform emerging negotiations concerning a wider package of climate change adaptation priorities with State and Federal governments. Next stage research priorities are also being developed to enable an ongoing alliance between researchers and the region’s climate change response.

Beyond point forecasting : evaluation of alternative prediction intervals for tourist arrivals

This paper evaluates the performances of prediction intervals generated from alternative time series models, in the context of tourism forecasting. The forecasting methods considered include the autoregressive (AR) model, the AR model using the bias-corrected bootstrap, seasonal ARIMA models, innovations state space models for exponential smoothing, and Harvey’s structural time series models. We use thirteen monthly time series for the number of tourist arrivals to Hong Kong and Australia. The mean coverage rates and widths of the alternative prediction intervals are evaluated in an empirical setting. It is found that all models produce satisfactory prediction intervals, except for the autoregressive model. In particular, those based on the biascorrected bootstrap perform best in general, providing tight intervals with accurate coverage rates, especially when the forecast horizon is long.

Plasma-assisted self-sharpening of platelet-structured single-crystalline carbon nanocones

A mechanism and model for the vertical growth of platelet-structured vertically aligned single-crystalline carbon nanostructures by the formation of graphene layers on a flat top surface are proposed and verified experimentally. It is demonstrated that plasma-related effects lead to self-sharpening of tapered nanocones to form needlelike nanostructures, in a good agreement with the predicted dependence of the radius of a nanocone's flat top on the incoming ion flux and surface temperature. The growth mechanism is relevant to a broad class of nanostructures including nanotips, nanoneedles, and nanowires and can be used to improve the predictability of nanofabrication processes. © 2007 American Institute of Physics.

Arterial short-term travel time prediction using Bluetooth data

This project recognized lack of data analysis and travel time prediction on arterials as the main gap in the current literature. For this purpose it first investigated reliability of data gathered by Bluetooth technology as a new cost effective method for data collection on arterial roads. Then by considering the similarity among varieties of daily travel time on different arterial routes, created a SARIMA model to predict future travel time values. Based on this research outcome, the created model can be applied for online short term travel time prediction in future.

Low-frequency surface waves in a structured magnetized dusty plasma

It is shown that charged dust particles carrying a considerable proportion of the negative charge of a structured magnetized plasma can lead to low-frequency electromagnetic surface waves which otherwise do not exist. The waves are Alfvén-like and propagate across the stationary external magnetic field with a frequency below the ion cyclotron but much above the dust cyclotron frequency. The dispersion characteristics of the modes are obtained and applications to space plasmas discussed. Copyright 1999 by the American Geophysical Union.

Enhanced stochastic mobility prediction with multi-output Gaussian processes

Outdoor robots such as planetary rovers must be able to navigate safely and reliably in order to successfully perform missions in remote or hostile environments. Mobility prediction is critical to achieving this goal due to the inherent control uncertainty faced by robots traversing natural terrain. We propose a novel algorithm for stochastic mobility prediction based on multi-output Gaussian process regression. Our algorithm considers the correlation between heading and distance uncertainty and provides a predictive model that can easily be exploited by motion planning algorithms. We evaluate our method experimentally and report results from over 30 trials in a Mars-analogue environment that demonstrate the effectiveness of our method and illustrate the importance of mobility prediction in navigating challenging terrain.

Prediction of thermal expansion properties of carbon nanotubes using molecular dynamics simulations

The axial coefficients of thermal expansion (CTE) of various carbon nanotubes (CNTs), i.e., single-wall carbon nanotubes (SWCNTs), and some multi-wall carbon nanotubes (MWCNTs), were predicted using molecular dynamics (MDs) simulations. The effects of two parameters, i.e., temperature and the CNT diameter, on CTE were investigated extensively. For all SWCNTs and MWCNTs, the obtained results clearly revealed that within a wide low temperature range, their axial CTEs are negative. As the diameter of CNTs decreases, this temperature range for negative axial CTEs becomes narrow, and positive axial CTEs appear in high temperature range. It was found that the axial CTEs vary nonlinearly with the temperature, however, they decrease linearly as the CNT diameter increases. Moreover, within a wide temperature range, a set of empirical formulations was proposed for evaluating the axial CTEs of armchair and zigzag SWCNTs using the above two parameters. Finally, it was found that the absolute value of the negative axial CTE of any MWCNT is much smaller than those of its constituent SWCNTs, and the average value of the CTEs of its constituent SWCNTs. The present fundamental study is very important for understanding the thermal behaviors of CNTs in such as nanocomposite temperature sensors, or nanoelectronics devices using CNTs.

Learned stochastic mobility prediction for planning with control uncertainty on unstructured terrain

Motion planning for planetary rovers must consider control uncertainty in order to maintain the safety of the platform during navigation. Modelling such control uncertainty is difficult due to the complex interaction between the platform and its environment. In this paper, we propose a motion planning approach whereby the outcome of control actions is learned from experience and represented statistically using a Gaussian process regression model. This mobility prediction model is trained using sample executions of motion primitives on representative terrain, and predicts the future outcome of control actions on similar terrain. Using Gaussian process regression allows us to exploit its inherent measure of prediction uncertainty in planning. We integrate mobility prediction into a Markov decision process framework and use dynamic programming to construct a control policy for navigation to a goal region in a terrain map built using an on-board depth sensor. We consider both rigid terrain, consisting of uneven ground, small rocks, and non-traversable rocks, and also deformable terrain. We introduce two methods for training the mobility prediction model from either proprioceptive or exteroceptive observations, and report results from nearly 300 experimental trials using a planetary rover platform in a Mars-analogue environment. Our results validate the approach and demonstrate the value of planning under uncertainty for safe and reliable navigation.