Approximate variational inference based on a finite sample of Gaussian latent variables

Variational methods are employed in situations where exact Bayesian inference becomes intractable due to the difficulty in performing certain integrals. Typically, variational methods postulate a tractable posterior and formulate a lower bound on the desired integral to be approximated, e.g. marginal likelihood. The lower bound is then optimised with respect to its free parameters, the so-called variational parameters. However, this is not always possible as for certain integrals it is very challenging (or tedious) to come up with a suitable lower bound. Here, we propose a simple scheme that overcomes some of the awkward cases where the usual variational treatment becomes difficult. The scheme relies on a rewriting of the lower bound on the model log-likelihood. We demonstrate the proposed scheme on a number of synthetic and real examples, as well as on a real geophysical model for which the standard variational approaches are inapplicable.     

Discriminating Conduct: Deploying Systems to Support Awareness in Organizations

Alongside the emergence of the use of fieldwork studies for design there has been a discussion on how best these studies can inform system development. Concerns have been expressed as to whether their most appropriate contribution is a list of requirements or design recommendations. This article explores a recurrent issue that has emerged from fieldwork studies in Computer-Supported Cooperative Work, awareness, and with respect to a particular system development project discusses some of the implications for the development and deployment of one particular kind of technology—image recognition systems—in particular, organizational settings. In the setting in question—surveillance centers or operations rooms—staff utilize a range of practices to maintain awareness. Rather than extending field studies so that they can better assist design, it may be considered how workplace studies can contribute to a respecification of key concepts, like awareness, that are critical to an understanding of how technologies are used and deployed in everyday environments.     

Efficiently adapting graphical models for selectivity estimation

Query optimizers rely on statistical models that succinctly describe the underlying data. Models are used to derive cardinality estimates for intermediate relations, which in turn guide the optimizer to choose the best query execution plan. The quality of the resulting plan is highly dependent on the accuracy of the statistical model that represents the data. It is well known that small errors in the model estimates propagate exponentially through joins, and may result in the choice of a highly sub-optimal query execution plan. Most commercial query optimizers make the attribute value independence assumption: all attributes are assumed to be statistically independent. This reduces the statistical model of the data to a collection of one-dimensional synopses (typically in the form of histograms), and it permits the optimizer to estimate the selectivity of a predicate conjunction as the product of the selectivities of the constituent predicates. However, this independence assumption is more often than not wrong, and is considered to be the most common cause of sub-optimal query execution plans chosen by modern query optimizers. We take a step towards a principled and practical approach to performing cardinality estimation without making the independence assumption. By carefully using concepts from the field of graphical models, we are able to factor the joint probability distribution over all the attributes in the database into small, usually two-dimensional distributions, without a significant loss in estimation accuracy. We show how to efficiently construct such a graphical model from the database using only two-way join queries, and we show how to perform selectivity estimation in a highly efficient manner. We integrate our algorithms into the PostgreSQL DBMS. Experimental results indicate that estimation errors can be greatly reduced, leading to orders of magnitude more efficient query execution plans in many cases. Optimization time is kept in the range of tens of milliseconds, making this a practical approach for industrial-strength query optimizers.     

Adoption and use of Java generics

Support for generic programming was added to the Java language in 2004, representing perhaps the most significant change to one of the most widely used programming languages today. Researchers and language designers anticipated this addition would relieve many long-standing problems plaguing developers, but surprisingly, no one has yet measured how generics have been adopted and used in practice. In this paper, we report on the first empirical investigation into how Java generics have been integrated into open source software by automatically mining the history of 40 popular open source Java programs, traversing more than 650 million lines of code in the process. We evaluate five hypotheses and research questions about how Java developers use generics. For example, our results suggest that generics sometimes reduce the number of type casts and that generics are usually adopted by a single champion in a project, rather than all committers. We also offer insights into why some features may be adopted sooner and others features may be held back.     

View-based propagator derivation

When implementing a propagator for a constraint, one must decide about variants: When implementing min, should one also implement max? Should one implement linear constraints both with unit and non-unit coefficients? Constraint variants are ubiquitous: implementing them requires considerable (if not prohibitive) effort and decreases maintainability, but will deliver better performance than resorting to constraint decomposition. This paper shows how to use views to derive propagator variants, combining the efficiency of dedicated propagator implementations with the simplicity and effortlessness of decomposition. A model for views and derived propagators is introduced. Derived propagators are proved to be perfect in that they inherit essential properties such as correctness and domain and bounds consistency. Techniques for systematically deriving propagators such as transformation, generalization, specialization, and type conversion are developed. The paper introduces an implementation architecture for views that is independent of the underlying constraint programming system. A detailed evaluation of views implemented in Gecode shows that derived propagators are efficient and that views often incur no overhead. Views have proven essential for implementing Gecode, substantially reducing the amount of code that needs to be written and maintained.     

Involving the user Through Library 2.0

Libraries, as we know them today, can be defined by the term Library 1.0. This defines the way resources are kept on shelves or at a computer behind a login. These resources can be taken from a shelf, checked out by the library staff, taken home for a certain length of time and absorbed, and then returned to the library for someone else to avail of. Library 1.0 is a one-directional service that takes people to the information they require. Library 2.0 – or L2 as it is now more commonly addressed as – aims to take the information to the people by bringing the library service to the Internet and getting the users more involved by encouraging feedback participation. This paper presents an overview of Library 2.0 and introduces web 2.0 concepts.     

A Secure Correspondent Router Protocol for NEMO Route Optimization

The Network Mobility (NEMO) protocol is needed to support the world-wide mobility of aircraft mobile networks across different access networks in the future IPv6 based aeronautical telecommunications network (ATN). NEMO suffers from the constraint that all traffic has to be routed via the home agent though. The already existing correspondent router (CR) protocol solves this triangular routing problem and permits to route packets on a direct path between the mobile network and the ground based correspondent nodes. We identify security deficiencies of this protocol that make it unsuitable for use within the ATN. We therefore propose a new route optimization procedure based on the CR protocol that provides a higher level of security. We evaluate our new protocol in three ways. We first conduct a simulation based handover performance study using an implementation of a realistic aeronautical access technology. We then investigate the mobility signaling overhead. Finally, we specify a threat model applicable for the aeronautical environment and use it to perform a security analysis of both the old and our new protocol. It is shown that our protocol is not only more secure but also provides better handover latency, smaller overhead in the aeronautical scenario and a higher level of resilience when compared to the original CR protocol.     

Motion Planning with Pulley, Rope, and Baskets

We study a motion planning problem where items have to be transported from the top room of a tower to the bottom of the tower, while simultaneously other items have to be transported in the opposite direction. Item sets are moved in two baskets hanging on a rope and pulley. To guarantee stability of the system, the weight difference between the contents of the two baskets must always stay below a given threshold. We prove that it is $\varPi_{2}^{p}$ -complete to decide whether some given initial situation of the underlying discrete system can lead to a given goal situation. Furthermore we identify several polynomially solvable special cases of this reachability problem, and we also settle the computational complexity of a number of related questions.     

Non-encapsulated thermo-liquid crystals for digital particle tracking thermography/velocimetry in microfluidics

The ever accelerating state of technology has powered an increasing interest in heat transfer solutions and process engineering innovations in the microfluidics domain. In order to carry out such developments, reliable heat transfer diagnostic techniques are necessary. Thermo-liquid crystal (TLC) thermography, in combination with particle image velocimetry, has been a widely accepted and commonly used technique for the simultaneous measurement and characterization of temperature and velocity fields in macroscopic fluid flows for several decades. However, low seeding density, volume illumination, and low TLC particle image quality at high magnifications present unsurpassed challenges to its application to three-dimensional flows with microscopic dimensions. In this work, a measurement technique to evaluate the color response of individual non-encapsulated TLC particles is presented. A Shirasu porous glass membrane emulsification approach was used to produce the non-encapsulated TLC particles with a narrow size distribution and a multi-variable calibration procedure, making use of all three RGB and HSI color components, as well as the proper orthogonally decomposed RGB components, was used to achieve unprecedented low uncertainty levels in the temperature estimation of individual particles, opening the door to simultaneous temperature and velocity tracking using 3D velocimetry techniques.