Managing climate change ‘by the numbers’ in a UK energy company

This paper examines how ideas and practices of accounting come together in turning the abstract concept of climate change into a new non-financial performance measure in a large energy company in the UK. It develops the notion of ‘governmental management’ to explain how the firm’s carbon dioxide emissions were transformed into a new organisational object that could be made quantifiable, measureable and ultimately manageable because of the modern power of accounting in tying disciplinary subjectivities and objectivities together whilst operating simultaneously at the level of individual and the organisation. Examining these interrelations highlights the constitutive nature of accounting in creating not just new categories for accounting’s attention, but in turn new organisational knowledge and knowledge experts in the making up accounting for climate change. Significantly, it appears these new knowledge experts are no longer accountants: which may help explain accounting’s evolution into evermore spheres of influence as we increasingly choose to manage our world ‘by the numbers’.

Learning the learning rate for prediction with expert advice

Most standard algorithms for prediction with expert advice depend on a parameter called the learning rate. This learning rate needs to be large enough to fit the data well, but small enough to prevent overfitting. For the exponential weights algorithm, a sequence of prior work has established theoretical guarantees for higher and higher data-dependent tunings of the learning rate, which allow for increasingly aggressive learning. But in practice such theoretical tunings often still perform worse (as measured by their regret) than ad hoc tuning with an even higher learning rate. To close the gap between theory and practice we introduce an approach to learn the learning rate. Up to a factor that is at most (poly)logarithmic in the number of experts and the inverse of the learning rate, our method performs as well as if we would know the empirically best learning rate from a large range that includes both conservative small values and values that are much higher than those for which formal guarantees were previously available. Our method employs a grid of learning rates, yet runs in linear time regardless of the size of the grid.

Efficient minimax strategies for square loss games

We consider online prediction problems where the loss between the prediction and the outcome is measured by the squared Euclidean distance and its generalization, the squared Mahalanobis distance. We derive the minimax solutions for the case where the prediction and action spaces are the simplex (this setup is sometimes called the Brier game) and the \ell_2 ball (this setup is related to Gaussian density estimation). We show that in both cases the value of each sub-game is a quadratic function of a simple statistic of the state, with coefficients that can be efficiently computed using an explicit recurrence relation. The resulting deterministic minimax strategy and randomized maximin strategy are linear functions of the statistic.

Drug accumulation into single drug-sensitive and drug-resistant prostate cancer cells conducted on the single cell bioanalyzer

Multidrug resistance (MDR) occurs in prostate cancer, and this happens when the cancer cells resist chemotherapeutic drugs by pumping them out of the cells. MDR inhibitors such as cyclosporin A (CsA) can stop the pumping and enhance the drugs accumulated in the cells. The cellular drug accumulation is monitored using a microfluidic chip mounted on a single cell bioanalyzer. This equipment has been developed to measure accumulation of drugs such as doxorubicin (DOX) and fluorescently labeled paclitaxel (PTX) in single prostate cancer cells. The inhibition of drug efflux on the same prostate cell was examined in drug-sensitive and drug-resistant cells. Accumulation of these drug molecules was not found in the MDR cells, PC-3 RX-DT2R cells. Enhanced drug accumulation was observed only after treating the MDR cell in the presence of 5 μM of CsA as the MDR inhibitor. We envision this monitoring of the accumulation of fluorescent molecules (drug or fluorescent molecules), if conducted on single patient cancer cells, can provide information for clinical monitoring of patients undergoing chemotherapy in the future.

Lock the gate: Fracking, investor-state dispute settlement, and the Trans-Pacific Partnership

There has been much debate about the relationship between international trade, the environment, biodiversity protection, and climate change.The Obama Administration has pushed such issues into sharp relief, with its advocacy for sweeping international trade agreements, such as the Trans-Pacific Partnership and the Trans-Atlantic Trade and Investment Partnership. There has been much public concern about the impact of the mega-trade deals upon the protection of the environment. In particular, there has been a debate about whether the Trans-Pacific Partnership will promote dirty fracking. Will the Trans-Pacific Partnership transform the Pacific Rim into a Gasland?There has been a particular focus upon investor-state dispute settlement being used by unconventional mining companies. Investor-state dispute settlement is a mechanism which enables foreign investors to seek compensation from national governments at international arbitration tribunals. In her prescient 2009 book, The Expropriation of Environmental Governance, Kyla Tienhaara foresaw the rise of investor-state dispute resolution of environmental matters. She observed:'Over the last decade there has been an explosive increase of cases investment arbitration. This is significant in terms of not only the number of disputes that have arisen and the number of states that have been involved, but also the novel types of dispute that have emerged. Rather than solely involving straightforward incidences of nationalization or breach of contract, modern disputes often revolve around public policy measures and implicate sensitive issues such as access to drinking water, development on sacred indigenous sites and the protection of biodiversity.'In her study, Kyla Tienhaara observed that investment agreements, foreign investment contracts and investment arbitration had significant implications for the protection for the protection of the environment. She concluded that arbitrators have made it clear that they can, and will, award compensation to investors that claim to have been harmed by environmental regulation. She also found that some of the cases suggest that the mere threat of arbitration is sufficient to chill environmental policy development. Tienhaara was equally concerned by the possibility that a government may use the threat of arbitration as an excuse or cover for its failure to improve environmental regulation. In her view, it is evident that arbitrators have expropriated certain fundamental aspects of environmental governance from states. Tienhaara held: As a result, environmental regulation has become riskier, more expensive, and less democratic, especially in developing countries. This article provides a comparative analysis of the battles over fracking, investment, trade, and the environment in a number of key jurisdictions including the United States, Canada, Australia, and New Zealand. Part 1 focuses upon the United States. Part 2 examines the dispute between the Lone Pine Resources Inc. and the Government of Canada over a fracking moratorium in Quebec. Part 3 charts the rise of the Lock the Gate Alliance in Australia, and its demands for a moratorium in respect of coal seam gas and unconventional mining. Part 4 focuses upon parallel developments in New Zealand. This article concludes that Pacific Rim countries should withdraw from investor-state dispute settlement procedures, because of the threat posed to environmental regulation in respect of air, land, and water.

Improving deep convultional neural networks with unsupervised feature learning

The latest generation of Deep Convolutional Neural Networks (DCNN) have dramatically advanced challenging computer vision tasks, especially in object detection and object classification, achieving state-of-the-art performance in several computer vision tasks including text recognition, sign recognition, face recognition and scene understanding. The depth of these supervised networks has enabled learning deeper and hierarchical representation of features. In parallel, unsupervised deep learning such as Convolutional Deep Belief Network (CDBN) has also achieved state-of-the-art in many computer vision tasks. However, there is very limited research on jointly exploiting the strength of these two approaches. In this paper, we investigate the learning capability of both methods. We compare the output of individual layers and show that many learnt filters and outputs of the corresponding level layer are almost similar for both approaches. Stacking the DCNN on top of unsupervised layers or replacing layers in the DCNN with the corresponding learnt layers in the CDBN can improve the recognition/classification accuracy and training computational expense. We demonstrate the validity of the proposal on ImageNet dataset.

On the biomechanical significance of inter-lamellar interfaces in the intervertebral disc

The intervertebral disc (IVD) is a unique soft tissue structure which provides structural support and flexibility in the axial skeleton of vertebrates. From a structural perspective, the disc behaves somewhat like a thick walled pressure vessel, where the walls are comprised of a series of composite annular rings (lamellae). However, a prior study (Marchand and Ahmed, 1990) found a high proportion of circumferentially discontinuous lamellae in human lumbar IVDs. The presence of these discontinuities raises important structural questions, because discontinuous lamellae cannot withstand high nucleus pressures via the generation of circumferential (hoop) stress. A possible alternative mechanism may be that inter-lamellar cohesion allows shear stress transfer between adjacent annular layers. The aim of the present study was therefore to investigate the importance of inter-lamellar shear resistance in the intervertebral disc. This work found that inter-lamellar shear resistance has a strong influence on the compressive stiffness of the intervertebral disc, with a change in interface condition from tied (no slip) to frictionless (no shear resistance) reducing disc compressive stiffness by 40%. However, it appears that substantial inter-lamellar shear resistance is present in the bovine tail disc. Decreases in inter-lamellar shear resistance due to degradation of bridging collagenous or elastic fibre structures could therefore be an important part of the process of disc degeneration.

Acceptance of prostitution and its social determinants in Canada

The nature of collective perception of prostitution is understudied in Canada. Except some rudimentary reports on the percentages of the key legal options, multivariate analysis has never been used to analyze the details of public opinion on prostitution. The current study explores the trend of public attitude toward prostitution acceptability in Canada over a 25-year span and examines the social determinants of the acceptability of prostitution, using structural equation modeling (SEM), which allows researchers to elaborate both direct and indirect effects (through mediating variables) on the outcome variable. Results show that the public has become more acceptant of prostitution over time. In addition, the less religious, less authoritarian, and more educated are more acceptant of prostitution than the more religious, more authoritarian, and less well educated. The effects of religiosity and authoritarianism mediate out the direct effects of age, gender, gender equality, marriage, marriage as an outdated institution, Quebec, race, and tolerance. The findings may serve as a reference point for the law reform regarding the regulation of prostitution in Canada.

Class-specific sparse codes for representing activities

In this paper we investigate the effectiveness of class specific sparse codes in the context of discriminative action classification. The bag-of-words representation is widely used in activity recognition to encode features, and although it yields state-of-the art performance with several feature descriptors it still suffers from large quantization errors and reduces the overall performance. Recently proposed sparse representation methods have been shown to effectively represent features as a linear combination of an over complete dictionary by minimizing the reconstruction error. In contrast to most of the sparse representation methods which focus on Sparse-Reconstruction based Classification (SRC), this paper focuses on a discriminative classification using a SVM by constructing class-specific sparse codes for motion and appearance separately. Experimental results demonstrates that separate motion and appearance specific sparse coefficients provide the most effective and discriminative representation for each class compared to a single class-specific sparse coefficients.

Combat sports analytics: Boxing punch classification using overhead depth imagery

In competitive combat sporting environments like boxing, the statistics on a boxer's performance, including the amount and type of punches thrown, provide a valuable source of data and feedback which is routinely used for coaching and performance improvement purposes. This paper presents a robust framework for the automatic classification of a boxer's punches. Overhead depth imagery is employed to alleviate challenges associated with occlusions, and robust body-part tracking is developed for the noisy time-of-flight sensors. Punch recognition is addressed through both a multi-class SVM and Random Forest classifiers. A coarse-to-fine hierarchical SVM classifier is presented based on prior knowledge of boxing punches. This framework has been applied to shadow boxing image sequences taken at the Australian Institute of Sport with 8 elite boxers. Results demonstrate the effectiveness of the proposed approach, with the hierarchical SVM classifier yielding a 96% accuracy, signifying its suitability for analysing athletes punches in boxing bouts.

Cues in the environment: A design principle for ambient intelligence

The aim of this paper is to propose design principles for ambient intelligence (AmI) environments. The question we are investigating is how these environments can be designed to support a group to be able to carry out common goal-oriented activities. The approach we are taking in answering this question is informed by the concept of collective intelligence (CI). We are applying the concept of CI to AmI as we have found it works well in biological and social systems. Examples from nature demonstrate the power of CI stimulated by implicit cues in the environment. We use these examples to derive design principles for AmI environments. By applying these design principles to a concrete scenario, we are able to propose ways to help decrease environmental pollution within urban areas.