Multiplicative Kernels: Object Detection, Segmentation and Pose Estimation

Object detection is challenging when the object class exhibits large within-class variations. In this work, we show that foreground-background classification (detection) and within-class classification of the foreground class (pose estimation) can be jointly learned in a multiplicative form of two kernel functions. One kernel measures similarity for foreground-background classification. The other kernel accounts for latent factors that control within-class variation and implicitly enables feature sharing among foreground training samples. Detector training can be accomplished via standard SVM learning. The resulting detectors are tuned to specific variations in the foreground class. They also serve to evaluate hypotheses of the foreground state. When the foreground parameters are provided in training, the detectors can also produce parameter estimate. When the foreground object masks are provided in training, the detectors can also produce object segmentation. The advantages of our method over past methods are demonstrated on data sets of human hands and vehicles.

Statistic Rate Monotonic Scheduling

In this paper we present Statistical Rate Monotonic Scheduling (SRMS), a generalization of the classical RMS results of Liu and Layland that allows scheduling periodic tasks with highly variable execution times and statistical QoS requirements. Similar to RMS, SRMS has two components: a feasibility test and a scheduling algorithm. The feasibility test for SRMS ensures that using SRMS' scheduling algorithms, it is possible for a given periodic task set to share a given resource (e.g. a processor, communication medium, switching device, etc.) in such a way that such sharing does not result in the violation of any of the periodic tasks QoS constraints. The SRMS scheduling algorithm incorporates a number of unique features. First, it allows for fixed priority scheduling that keeps the tasks' value (or importance) independent of their periods. Second, it allows for job admission control, which allows the rejection of jobs that are not guaranteed to finish by their deadlines as soon as they are released, thus enabling the system to take necessary compensating actions. Also, admission control allows the preservation of resources since no time is spent on jobs that will miss their deadlines anyway. Third, SRMS integrates reservation-based and best-effort resource scheduling seamlessly. Reservation-based scheduling ensures the delivery of the minimal requested QoS; best-effort scheduling ensures that unused, reserved bandwidth is not wasted, but rather used to improve QoS further. Fourth, SRMS allows a system to deal gracefully with overload conditions by ensuring a fair deterioration in QoS across all tasks---as opposed to penalizing tasks with longer periods, for example. Finally, SRMS has the added advantage that its schedulability test is simple and its scheduling algorithm has a constant overhead in the sense that the complexity of the scheduler is not dependent on the number of the tasks in the system. We have evaluated SRMS against a number of alternative scheduling algorithms suggested in the literature (e.g. RMS and slack stealing), as well as refinements thereof, which we describe in this paper. Consistently throughout our experiments, SRMS provided the best performance. In addition, to evaluate the optimality of SRMS, we have compared it to an inefficient, yet optimal scheduler for task sets with harmonic periods.

Research and design of high-speed advanced analogue front-ends for fibre-optic transmission systems

In the last decade, we have witnessed the emergence of large, warehouse-scale data centres which have enabled new internet-based software applications such as cloud computing, search engines, social media, e-government etc. Such data centres consist of large collections of servers interconnected using short-reach (reach up to a few hundred meters) optical interconnect. Today, transceivers for these applications achieve up to 100Gb/s by multiplexing 10x 10Gb/s or 4x 25Gb/s channels. In the near future however, data centre operators have expressed a need for optical links which can support 400Gb/s up to 1Tb/s. The crucial challenge is to achieve this in the same footprint (same transceiver module) and with similar power consumption as today’s technology. Straightforward scaling of the currently used space or wavelength division multiplexing may be difficult to achieve: indeed a 1Tb/s transceiver would require integration of 40 VCSELs (vertical cavity surface emitting laser diode, widely used for short‐reach optical interconnect), 40 photodiodes and the electronics operating at 25Gb/s in the same module as today’s 100Gb/s transceiver. Pushing the bit rate on such links beyond today’s commercially available 100Gb/s/fibre will require new generations of VCSELs and their driver and receiver electronics. This work looks into a number of state‐of-the-art technologies and investigates their performance restraints and recommends different set of designs, specifically targeting multilevel modulation formats. Several methods to extend the bandwidth using deep submicron (65nm and 28nm) CMOS technology are explored in this work, while also maintaining a focus upon reducing power consumption and chip area. The techniques used were pre-emphasis in rising and falling edges of the signal and bandwidth extensions by inductive peaking and different local feedback techniques. These techniques have been applied to a transmitter and receiver developed for advanced modulation formats such as PAM-4 (4 level pulse amplitude modulation). Such modulation format can increase the throughput per individual channel, which helps to overcome the challenges mentioned above to realize 400Gb/s to 1Tb/s transceivers.

Practice links [Issue 58, June 2014]

Practice Links is a free e-publication for practitioners working in Irish social services, voluntary and nongovernmental sectors. Practice Links was created to enable practitioners to keep up-to-date with new publications, electronic resources and conference opportunities. Issue 58 includes listings for upcoming conferences, resources, recent policy reports as well as reviews of publications.

Uncle Tom's Cabin and the Irish National Tale

Accepted Version

Glutamine supplementation and renal health

Gemstone Team Juiced

New data inversion formulae in confocal scanning microscopy

Whereas the resolving power of an ordinary optical microscope is determined by the classical Rayleigh distance, significant super-resolution, i.e. resolution improvement beyond that Rayleigh limit, has been achieved by confocal scanning light microscopy. Furthermore is has been shown that the resolution of a confocal scanning microscope can still be significantly enhanced by measuring, for each scanning position, the full diffraction image by means of an array of detectors and by inverting these data to recover the value of the object at the focus. We discuss the associated inverse problem and show how to generalize the data inversion procedure by allowing, for reconstructing the object at a given point, to make use also of the diffraction images recorded at other scanning positions. This leads us to a whole family of generalized inversion formulae, which contains as special cases some previously known formulae. We also show how these exact inversion formulae can be implemented in practice.

De nietos y aves

Soluciones a los Problemas de los abuelos, estudio de las respuestas a uno de los problemas planteados en el Torneo de Matemáticas para 2º de la ESO de la Sociedad Canaria Isaac Newton de profesores de matemáticas. Actividad de resolución de problemas en una clase de 6º de Primaria.

Las prácticas sociales de modelación en la construcción de lo exponencial

La intención de la ponencia está en la dirección de presentar un estudio de las prácticas que ejercen los actores en un diseño de aprendizaje puesto en escena en el aula de matemáticas. El diseño referido se centra, no en los contenidos matemáticos en sí o en las producciones de los participantes, sino en las prácticas sociales ejercidas por los participantes utilizando herramientas y situadas en un contexto social; en este caso las prácticas sociales de modelación del enfriamiento de un líquido. Reportamos la narración de la puesta en escena en el aula de matemáticas de un diseño de aprendizaje basado en prácticas sociales de modelación de fenómenos: “Lo exponencial: la ley de enfriamiento de Newton”. Aquí narramos como los participantes construyen lo exponencial como herramienta al intentar comprender y predecir lo que sucede al enfriarse un líquido.

El problema de basilea. El año de Euler: 1707-2007

El pasado 15 de abril se cumplían 300 años del nacimiento de uno de los cuatro matemáticos más geniales de la historia, Leonhard Euler. Para mí, los otros tres, y que cada cual elija su orden, son Arquímedes, Newton y Gauss. Si la calificación la hiciésemos atendiendo a la cantidad de los trabajos de primer orden realizados por cada uno de ellos, sin duda Euler ocuparía el primer lugar. A lo largo de su extensa vida Euler produjo más de ochocientos libros y miles de artículos y trabajos. Sus obras completas Opera Omnia ocupan más de 80 volúmenes. Sin lugar a dudas es el matemático más prolífico de la Historia. Pero, con ser importante la cantidad de trabajos, el aprecio de los matemáticos contemporáneos y posteriores a él se debe más a la riqueza, originalidad, belleza y genial agudeza de su obra que a su volumen.

A defect equation approach for the coupling of subdomains in domain decomposition methods

A defect equation for the coupling of nonlinear subproblems defined in nonoverlapped subdomains arise in domain decomposition methods is presented. Numerical solutions of defect equations by means of quasi-Newton methods are considered.

Flexible self-adjustment of task deployment in dynamic environments

This paper presents an investigation into dynamic self-adjustment of task deployment and other aspects of self-management, through the embedding of multiple policies. Non-dedicated loosely-coupled computing environments, such as clusters and grids are increasingly popular platforms for parallel processing. These abundant systems are highly dynamic environments in which many sources of variability affect the run-time efficiency of tasks. The dynamism is exacerbated by the incorporation of mobile devices and wireless communication. This paper proposes an adaptive strategy for the flexible run-time deployment of tasks; to continuously maintain efficiency despite the environmental variability. The strategy centres on policy-based scheduling which is informed by contextual and environmental inputs such as variance in the round-trip communication time between a client and its workers and the effective processing performance of each worker. A self-management framework has been implemented for evaluation purposes. The framework integrates several policy-controlled, adaptive services with the application code, enabling the run-time behaviour to be adapted to contextual and environmental conditions. Using this framework, an exemplar self-managing parallel application is implemented and used to investigate the extent of the benefits of the strategy