Time filter

Source Type

Sheppard C.J.R.,National University of Singapore | Sheppard C.J.R.,Singapore MIT Alliance for Research and Technology SMART
Optics Letters | Year: 2012

Recently, we described a geometric construction for determining the eigenvalues of the coherency matrix in three dimensions. We show that this leads directly to a representation of the three-dimensional degree of polarization in terms of a triangular composition plot, in which different polarization measures have simple properties and can be expressed in terms of the matrix invariants. This composition plot is an alternative to the spherical plot recently used to illustrate the degree of polarization in terms of entanglement. © 2012 Optical Society of America.

Sheppard C.J.R.,National University of Singapore | Sheppard C.J.R.,Singapore MIT Alliance for Research and Technology SMART
Journal of the Optical Society of America A: Optics and Image Science, and Vision | Year: 2011

Various different parameters have been introduced to describe the degree of polarization of a partially polarized electromagnetic field in three dimensions. Of these, parameters based on the eigenvalues of the coherency matrix are invariant under a unitary transformation. Here, explicit expressions are presented for the eigenvalues, thus providing a geometrical interpretation of the behavior. These expressions are applied to the Huynen decomposition and allow interrelations between different parameters to be developed. © 2011 Optical Society of America.

Rodrigues F.,University of Coimbra | Pereira F.,Singapore MIT Alliance for Research and Technology SMART | Ribeiro B.,University of Coimbra
Pattern Recognition Letters | Year: 2013

With the increasing popularity of online crowdsourcing platforms such as Amazon Mechanical Turk (AMT), building supervised learning models for datasets with multiple annotators is receiving an increasing attention from researchers. These platforms provide an inexpensive and accessible resource that can be used to obtain labeled data, and in many situations the quality of the labels competes directly with those of experts. For such reasons, much attention has recently been given to annotator-aware models. In this paper, we propose a new probabilistic model for supervised learning with multiple annotators where the reliability of the different annotators is treated as a latent variable. We empirically show that this model is able to achieve state of the art performance, while reducing the number of model parameters, thus avoiding a potential overfitting. Furthermore, the proposed model is easier to implement and extend to other classes of learning problems such as sequence labeling tasks. © 2013 Elsevier B.V. All rights reserved.

Li R.,Massachusetts Institute of Technology | Li R.,Tsinghua University | Pereira F.C.,Singapore MIT Alliance for Research and Technology SMART | Ben-Akiva M.E.,Massachusetts Institute of Technology
Accident Analysis and Prevention | Year: 2014

Traffic incident duration is known to result from a combination of multiple factors, including covariates such as spatial and temporal characteristics, traffic conditions, and existence of secondary accidents but also the clearance method itself. In this paper, a competing risks mixture model is used to investigate the influence of clearance methods and various covariates on the duration of traffic incidents and predict traffic incident duration. The proposed mixture model considers the uncertainty in any of five clearance methods that occurred. The probability of the clearance method is specified in the mixture by using a multinomial logistic model. Three candidate distributions, namely, generalized gamma, Weibull, and log-logistic are tested to determine the most appropriate probability density function of the parametric survival analysis model. The unobserved heterogeneity is also incorporated into the mixture model in a way that allows parameters to vary across observations based on the three candidate distributions. The methods are illustrated with incident data from Singaporean expressways from January 2010 to December 2011. Regression analysis reveals that the probability of different clearance methods and the duration of traffic incidents are both significantly affected by various factors, such as traffic conditions and incident characteristics. Results show that the proposed mixture model is better than the traditional accelerated failure time model, and it predicts traffic incident duration with reasonable accuracy, as shown by the mean average percent error. © 2014 Elsevier Ltd. All rights reserved.

Velasco E.,Singapore MIT Alliance for Research and Technology SMART | Roth M.,National University of Singapore
Journal of the Air and Waste Management Association | Year: 2012

Singapore has many environmental accomplishments to its credit. Accessible data on air quality indicates that all criteria pollutants satisfy both U.S. Environmental Protection Agency (EPA) and World Health Organization (WHO) air quality standards and guidelines, respectively. The exception is PM2.5 (particles with an aerodynamic diameter ≤2.5 μm), which is not currently considered a criteria pollutant in Singapore but may potentially be the major local air pollution problem and cause for health concern. Levels of other airborne pollutants as well as their physical and chemical processes associated with local formation, transformation, dispersion, and deposition are not known. According to available emission inventories, Singapore's contribution to the total atmospheric pollution and carbon budget at the regional and global scales is small. Emissions per unit gross domestic product (GDP) are low compared with other countries, although Singapore's per-capita GDP and per-capita emissions are among the highest in the world. Some information is available on health effects, but the impacts on the ecosystem and the complex interactions of air pollution and climate change at a regional level are also unknown. This article reviews existing available information on atmospheric pollution and greenhouse gas emissions and proposes a multipollutant approach to greenhouse gas mitigation and local air quality. Singapore, by reducing its per-capita emissions, increasing the availability of information (e.g., through regularly publishing hourly and/or daily PM2.5 concentrations) and developing a research agenda in this area, would likely be seen to be a model of a high-density, livable, and sustainable city in Southeast Asia and other tropical regions worldwide. Singapore is widely recognized for its environmental achievements and often cited as a model of a high-density, livable, and sustainable city. This article reviews available information with the aim to provide a reference for future scientific research of strategic relevance for Singapore's air quality and greenhouse gas mitigation management under a multipollutant framework. However, the limited publicly accessible data and little scientific information prevent a comprehensive assessment of the local air quality and greenhouse gas emissions. Singapore's dynamic economy and strong profile in advanced science and technological innovation have the potential to enhance the research agenda in this area, which is not yet well developed in tropical cities. © Copyright 2012 A&WMA.

Discover hidden collaborations