Methods to accelerate the learning of bayesian network structures

R. Daly and Q. Shen. Methods to accelerate the learning of bayesian network structures. Proceedings of the Proceedings of the 2007 UK Workshop on Computational Intelligence.

Using ant colony optimisation in learning Bayesian network equivalence classes

R. Daly, Q. Shen and S. Aitken. Using ant colony optimisation in learning Bayesian network equivalence classes. Proceedings of the 2006 UK Workshop on Computational Intelligence, pages 111-118.

Fault identification through the combination of symbolic conflict recognition and Markov Chain-aided belief revision

F. Smith and Q. Shen. Fault identification through the combination of symbolic conflict recognition and Markov Chain-aided belief revision. IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans, 34(5):649-663, 2004.

I Cannot Tell: Edward Thomas's Uncertainties

Poster, Jeremy, 'I Cannot Tell: Edward Thomas's Uncertainties', In: 'Branch-Lines: Edward Thomas and Contemporary Poetry', Guy Cuthbertson & Lucy Newlyn (eds), (London: Enitharmon Press), pp.264, 2007 RAE2008

Search/ing for missing people: families living with ambiguous absence

Families of missing people are often understood as inhabiting a particular space of ambiguity, captured in the phrase ‘living in limbo’ (Holmes, 2008). To explore this uncertain ground, we interviewed 25 family members to consider how human absence is acted upon and not just felt within this space ‘in between’ grief and loss (Wayland, 2007). In the paper, we represent families as active agents in spatial stories of ‘living in limbo’, and we provide insights into the diverse strategies of search/ing (technical, physical and emotional) in which they engage to locate either their missing member or news of them. Responses to absence are shown to be intimately bound up with unstable spatial knowledges of the missing person and emotional actions that are subject to change over time. We suggest that practices of search are not just locative actions, but act as transformative processes providing insights into how families inhabit emotional dynamism and transition in response to the on-going ‘missing situation’ and ambiguous loss (Boss, 1999, 2013).

O wielomianach cyklotomicznych

Wydział Matematyki i Informatyki

Napięcia tożsamościowe, tematyczne i formalne w dziele Yasminy Khadry

Wydział Neofilologii Instytut Filologii Romańskiej

Resíduos de medicamentos

Projeto de Pós-Graduação/Dissertação apresentado à Universidade Fernando Pessoa como parte dos requisitos para obtenção do grau de Mestre em Ciências Farmacêuticas

Simplified decoding techniques for linear block codes

Error correcting codes are combinatorial objects, designed to enable reliable transmission of digital data over noisy channels. They are ubiquitously used in communication, data storage etc. Error correction allows reconstruction of the original data from received word. The classical decoding algorithms are constrained to output just one codeword. However, in the late 50’s researchers proposed a relaxed error correction model for potentially large error rates known as list decoding. The research presented in this thesis focuses on reducing the computational effort and enhancing the efficiency of decoding algorithms for several codes from algorithmic as well as architectural standpoint. The codes in consideration are linear block codes closely related to Reed Solomon (RS) codes. A high speed low complexity algorithm and architecture are presented for encoding and decoding RS codes based on evaluation. The implementation results show that the hardware resources and the total execution time are significantly reduced as compared to the classical decoder. The evaluation based encoding and decoding schemes are modified and extended for shortened RS codes and software implementation shows substantial reduction in memory footprint at the expense of latency. Hermitian codes can be seen as concatenated RS codes and are much longer than RS codes over the same aphabet. A fast, novel and efficient VLSI architecture for Hermitian codes is proposed based on interpolation decoding. The proposed architecture is proven to have better than Kötter’s decoder for high rate codes. The thesis work also explores a method of constructing optimal codes by computing the subfield subcodes of Generalized Toric (GT) codes that is a natural extension of RS codes over several dimensions. The polynomial generators or evaluation polynomials for subfield-subcodes of GT codes are identified based on which dimension and bound for the minimum distance are computed. The algebraic structure for the polynomials evaluating to subfield is used to simplify the list decoding algorithm for BCH codes. Finally, an efficient and novel approach is proposed for exploiting powerful codes having complex decoding but simple encoding scheme (comparable to RS codes) for multihop wireless sensor network (WSN) applications.

Reflections on the transformation of my meaning making to support the organic growth of a firm

The objective of my Portfolio is to explore the working hypothesis that the organic growth of a firm is governed by the perspectives of individuals and such perspectives are governed by their meaning-making. The Portfolio presents explorations of the transformation of my meaning making and in adopting new practices to support the organic growth of a firm. I use the work of other theorists to transition my understanding of how the world works. This transition process is an essential tool to engage with and understand the perspectives of others and develop a mental capacity to “train one’s imagination to go visiting” (Arendt, 1982; p.43). The Portfolio, therefore, is primarily located in reflective research. Using Kegan’s (1994) approach to Adult Mental Development, and Sowell’s (2007) understanding of the visions which silently shape our thoughts I organise the developments of my meaning making around three transformation pillars of change. In pillar one I seek to transform an unthinking respect for authority and break down a blind pervasiveness of thought within my reasoning process arising from an instinct for attachment and support from others whom I trust. In pillar two I seek to discontinue using autocratic leadership and learn to use the thoughts and contributions of a wider team to make improved choices about uncertain future events. In pillar three I explore the use of a more reflective thinking framework to test the accuracy of my perceptions and apply a high level of integrity in my reasoning process. The transformation of my meaning making has changed my perspectives and in turn my preferred practices to support the organic growth of a firm. I identify from practice that a transformative form of leadership is far more effective that a transactional form of leadership to stimulate the trust and teamwork required to sustain the growth a firm. Creating an environment where one feels free to share thoughts and feelings with others is an essential tool to build a team to critique the thoughts of one other. Furthermore, the entrepreneurial wisdom to grow a firm must come from a wider team, located both inside and outside the boundaries of a firm. No individual or small team has the mental capacity to provide the entrepreneurship required to drive the organic growth of a firm. I address my Portfolio to leaders in organisations who have no considered framework on the best practices required to lead a social organisation. These individuals may have no sense of what they implicitly believe drives social causation and they may have no understanding if their meaning making supports or curtails the practices required to grow a firm. They may have a very limited capacity to think in a logical manner, with the result they are using guesses from their ‘gut’ to make poor judgements in the management of a firm.

Chamber investigations of atmospheric mercury oxidation chemistry

Mercury is a potent neurotoxin even at low concentrations. The unoxidised metal has a high vapour pressure and can circulate through the atmosphere, but when oxidised can deposit and be accumulated through the food chain. This work aims to investigate the oxidation processes of atmospheric Hg0(g). The first part describes efforts to make a portable Hg sensor based on Cavity Enhanced Absorption Spectroscopy (CEAS). The detection limit achieved was 66 ngm−3 for a 10 second averaging time. The second part of this work describes experiments carried out in a temperature controlled atmospheric simulation chamber in the Desert Research Institute, Reno, Nevada, USA. The chamber was built around an existing Hg CRDS system that could measure Hg concentrations in the chamber of<100 ngm−3 at 1 Hz enabling reactions to be followed. The main oxidant studied was bromine, which was quantified with a LED based CEAS system across the chamber. Hg oxidation in the chamber was found to be mostly too slow for current models to explain. A seven reaction model was developed and tested to find which parameters were capable of explaining the deviation. The model was overdetermined and no unique solution could be found. The most likely possibility was that the first oxidation step Hg + Br →HgBr was slower than the preferred literature value by a factor of two. However, if the more uncertain data at low [Br2] was included then the only parameter that could explain the experiments was a fast, temperature independent dissociation of HgBr some hundreds of times faster than predicted thermolysis or photolysis rates. Overall this work concluded that to quantitatively understand the reaction of Hg with Br2, the intermediates HgBr and Br must be measured. This conclusion will help to guide the planning of future studies of atmospheric Hg chemistry.

Use of pulsed field gel electrophoresis to characterize BCR gene involvement in CML patients lacking M-BCR rearrangement.

We studied the pattern of BCR involvement in 52 patients with chronic myeloid leukemia by Southern blotting. Of 33 Philadelphia (Ph)-positive patients, 30 had evidence of M-BCR rearrangement, two cases were difficult to interpret, and one clearly lacked evidence of M-BCR rearrangement. Of 19 Ph-negative patients, nine showed M-BCR rearrangement, nine showed no rearrangement, and one result was uncertain. We selected for more detailed study eight patients (three Ph-positive and five Ph-negative). Two of the Ph-positive patients, whose Southern blots were difficult to interpret, had rearranged bands when the BCR gene was studied by pulsed field gel electrophoresis (PFGE). Results of PFGE studies and in situ hybridization to metaphase chromosomes in the third Ph-positive patient, whose DNA clearly lacked M-BCR rearrangement on Southern analysis, were consistent with a breakpoint on chromosome 22 located 3' of all known exons of the BCR gene. However, mRNA studied with the polymerase chain reaction showed evidence of a classical b2-a2 linkage. The findings in this patient may be explained by an unusual genomic breakpoint downstream of the BCR gene associated with long range splicing that excluded all of the 3' BCR exons. Of the five patients with Ph-negative M-BCR non-rearranged CML studied by PFGE for BCR gene rearrangement, none had evidence of rearranged bands. We conclude that PFGE is a valuable adjunct to standard molecular techniques for the study of atypical cases of CML. Occasional patients with Ph-positive CML have breakpoints outside M-BCR. The BCR gene is probably not involved in patients with Ph-negative, M-BCR non-rearranged CML.

THE EFFECT OF A PROGRAM OF PORTABLE ELECTRONIC PIANO KEYBOARD EXPERIENCE ON THE ACQUISITION OF SIGHT-SINGING SKILL IN THE NOVICE HIGH SCHOOL CHORUS

The problem was to determine whether a method of aural and visual vocal training that included a program of portable electronic piano keyboard experience would be more effective in teaching sight-singing skills to novice high school chorus students than a method that included only aural and visual vocal training. A sub-problem was to determine whether novice chorus students enjoyed playing electronic keyboards in chorus as a reinforcement experience in sight-singing training. Students were randomly assigned to two treatment groups, tested with the Musical Aptitude Profile, Tonal Imagery, part A, and then trained separately. The experimental group sang repetitions of melodic patterns and utilized techniques associated with the Kodály Method while simultaneously playing keyboard. The comparison group received a similar treatment without using keyboards. The students were pre- and post-tested in sight-singing using the Vocal Sight-Reading Inventory. Results of the Analysis of Covariance using MAP scores as the covariate revealed no significant difference (p<.05) between post-test scores of the two groups. Improvement was noted in 96% of students from pre-test to post-test regardless of grouping. The repeated measures ANOVA revealed a significant relationship (p<.006) between aptitude group and post-test score. High aptitude students in both groups were found to benefit more from the training than low aptitude students. High aptitude keyboard group students achieved an average gain score that was 8.67 points higher than the comparison group. Of the total experimental group, 92% enjoyed playing keyboards in chorus. It is recommended that future research be undertaken to study the use of keyboards with advanced high school choruses and with uncertain singers in the high school chorus. Research is also needed to develop graded, valid, and reliable sight-singing tests for use in high school chorus. Techniques of the Kodály Method should be further investigated for use in high school sight-singing training.

Adaptive Mixture Modelling Metropolis Methods for Bayesian Analysis of Non-linear State-Space Models.

We describe a strategy for Markov chain Monte Carlo analysis of non-linear, non-Gaussian state-space models involving batch analysis for inference on dynamic, latent state variables and fixed model parameters. The key innovation is a Metropolis-Hastings method for the time series of state variables based on sequential approximation of filtering and smoothing densities using normal mixtures. These mixtures are propagated through the non-linearities using an accurate, local mixture approximation method, and we use a regenerating procedure to deal with potential degeneracy of mixture components. This provides accurate, direct approximations to sequential filtering and retrospective smoothing distributions, and hence a useful construction of global Metropolis proposal distributions for simulation of posteriors for the set of states. This analysis is embedded within a Gibbs sampler to include uncertain fixed parameters. We give an example motivated by an application in systems biology. Supplemental materials provide an example based on a stochastic volatility model as well as MATLAB code.

Physical and economic potential of geological CO2 storage in saline aquifers.

Carbon sequestration in sandstone saline reservoirs holds great potential for mitigating climate change, but its storage potential and cost per ton of avoided CO2 emissions are uncertain. We develop a general model to determine the maximum theoretical constraints on both storage potential and injection rate and use it to characterize the economic viability of geosequestration in sandstone saline aquifers. When applied to a representative set of aquifer characteristics, the model yields results that compare favorably with pilot projects currently underway. Over a range of reservoir properties, maximum effective storage peaks at an optimal depth of 1600 m, at which point 0.18-0.31 metric tons can be stored per cubic meter of bulk volume of reservoir. Maximum modeled injection rates predict minima for storage costs in a typical basin in the range of $2-7/ ton CO2 (2005 U.S.$) depending on depth and basin characteristics in our base-case scenario. Because the properties of natural reservoirs in the United States vary substantially, storage costs could in some cases be lower or higher by orders of magnitude. We conclude that available geosequestration capacity exhibits a wide range of technological and economic attractiveness. Like traditional projects in the extractive industries, geosequestration capacity should be exploited starting with the low-cost storage options first then moving gradually up the supply curve.