Solving Limited Memory Influence Diagrams

We present a new algorithm for exactly solving decision making problems represented as influence diagrams. We do not require the usual assumptions of no forgetting and regularity; this allows us to solve problems with simultaneous decisions and limited information. The algorithm is empirically shown to outperform a state-of-the-art algorithm on randomly generated problems of up to 150 variables and 10^64 solutions. We show that the problem is NP-hard even if the underlying graph structure of the problem has small treewidth and the variables take on a bounded number of states, but that a fully polynomial time approximation scheme exists for these cases. Moreover, we show that the bound on the number of states is a necessary condition for any efficient approximation scheme.

Some current problems in perovskite nano-ferroelectrics and multiferroics: kinetically-limited systems of finite lateral size

We describe some unsolved problems of current interest; these involve quantum critical points in
ferroelectrics and problems which are not amenable to the usual density functional theory, nor to
classical Landau free energy approaches (they are kinetically limited), nor even to the Landau–
Kittel relationship for domain size (they do not satisfy the assumption of infinite lateral diameter)
because they are dominated by finite aperiodic boundary conditions.

Robust bimodal person identification using face and speech with limited training data and corruption of both modalities

This paper presents a novel method of audio-visual fusion for person identification where both the speech and facial modalities may be corrupted, and there is a lack of prior knowledge about the corruption. Furthermore, we assume there is a limited amount of training data for each modality (e.g., a short training speech segment and a single training facial image for each person). A new representation and a modified cosine similarity are introduced for combining and comparing bimodal features with limited training data as well as vastly differing data rates and feature sizes. Optimal feature selection and multicondition training are used to reduce the mismatch between training and testing, thereby making the system robust to unknown bimodal corruption. Experiments have been carried out on a bimodal data set created from the SPIDRE and AR databases with variable noise corruption of speech and occlusion in the face images. The new method has demonstrated improved recognition accuracy.

The paradox of invasion: Reeves’ muntjac deer invade the British Isles from a limited number of founding females

High levels of genetic diversity and high propagule pressure are favoured by conservation biologists as the basis for successful reintroductions and ensuring the persistence of populations. However, invasion ecologists recognize the ‘paradox of invasion’, as successful species introductions may often be characterized by limited numbers of individuals and associated genetic bottlenecks. In the present study, we used a combination of high-resolution nuclear and mitochondrial genetic markers to investigate the invasion history of Reeves' muntjac deer in the British Isles. This invasion has caused severe economic and ecological damage, with secondary spread currently a concern throughout Europe and potentially globally. Microsatellite analysis based on eight loci grouped all 176 introduced individuals studied from across the species' range in the UK into one genetic cluster, and seven mitochondrial D-loop haplotypes were recovered, two of which were present at very low frequency and were related to more common haplotypes. Our results indicate that the entire invasion can be traced to a single founding event involving a low number of females. These findings highlight the fact that even small releases of species may, if ignored, result in irreversible and costly invasion, regardless of initial genetic diversity or continual genetic influx.

Teamwork Efficiency and Company Size

We study how ownership structure and management objectives interact in determining the company size without assuming information constraints or any explicit costs of management. In symmetric agent economies, the optimal company size balances the returns to scale of the production function and the returns to collaboration efficiency. For a general class of payoff functions, we characterize the optimal company size, and we compare the optimal company size across different managerial objectives. We demonstrate the restrictiveness of common assumptions on effort aggregation (e.g., constant elasticity of effort substitution), and we show that common intuition (e.g., that corporate companies are more efficient and therefore will be larger than equal-share partnerships) might not hold in general.

On Variable Selection of a Nonlinear Non-parametric System with a Limited Data Set: A Stepwise Algorithm

his paper considers a problem of identification for a high dimensional nonlinear non-parametric system when only a limited data set is available. The algorithms are proposed for this purpose which exploit the relationship between the input variables and the output and further the inter-dependence of input variables so that the importance of the input variables can be established. A key to these algorithms is the non-parametric two stage input selection algorithm.

Review of Slavery by Any Other Name: African Life under Company Rule in Colonial Mozambique.

Book review of Slavery by Any Other Name: African Life under Company Rule in Colonial Mozambique, by Eric Allina, Charlottesville, University of Virginia Press, 2012, 255 pp., £44.50, ISBN 978-0-8139-3272-9.

Limited range coverage problems

Tal como o título indica, esta tese estuda problemas de cobertura com alcance limitado. Dado um conjunto de antenas (ou qualquer outro dispositivo sem fios capaz de receber ou transmitir sinais), o objectivo deste trabalho é calcular o alcance mínimo das antenas de modo a que estas cubram completamente um caminho entre dois pontos numa região. Um caminho que apresente estas características é um itinerário seguro. A definição de cobertura é variável e depende da aplicação a que se destina. No caso de situações críticas como o controlo de fogos ou cenários militares, a definição de cobertura recorre à utilização de mais do que uma antena para aumentar a eficácia deste tipo de vigilância. No entanto, o alcance das antenas deverá ser minimizado de modo a manter a vigilância activa o maior tempo possível. Consequentemente, esta tese está centrada na resolução deste problema de optimização e na obtenção de uma solução particular para cada caso. Embora este problema de optimização tenha sido investigado como um problema de cobertura, é possível estabelecer um paralelismo entre problemas de cobertura e problemas de iluminação e vigilância, que são habitualmente designados como problemas da Galeria de Arte. Para converter um problema de cobertura num de iluminação basta considerar um conjunto de luzes em vez de um conjunto de antenas e submetê-lo a restrições idênticas. O principal tema do conjunto de problemas da Galeria de Arte abordado nesta tese é a 1-boa iluminação. Diz-se que um objecto está 1-bem iluminado por um conjunto de luzes se o invólucro convexo destas contém o objecto, tornando assim este conceito num tipo de iluminação de qualidade. O objectivo desta parte do trabalho é então minimizar o alcance das luzes de modo a manter uma iluminação de qualidade. São também apresentadas duas variantes da 1-boa iluminação: a iluminação ortogonal e a boa !-iluminação. Esta última tem aplicações em problemas de profundidade e visualização de dados, temas que são frequentemente abordados em estatística. A resolução destes problemas usando o diagrama de Voronoi Envolvente (uma variante do diagrama de Voronoi adaptada a problemas de boa iluminação) é também proposta nesta tese.

Curricular training report: pharmaceutical industry & spin-off company

O presente relatório expõe as atividades desenvolvidas durante o estágio curricular, frequentado na Bluepharma - Indústria Farmacêutica S.A. e na TREAT U, Lda. uma Spin-off da Universidade de Coimbra, no âmbito do Mestrado em Biomedicina Farmacêutica da Universidade de Aveiro. Esta foi uma experiência de 6 meses que teve duas componentes, uma multidisciplinar e outra monodisciplinar, as quais me permitiram desenvolver os conhecimentos e aptidões adquiridas ao longo do curso de mestrado e de as aplicar ao mundo real. Para além do desenvolvimento de competências profissionais, esta experiência possibilitou também a aquisição e desenvolvimento de várias aptidões, quer a nível pessoal como social. Nos primeiros dois meses desta minha experiência adquiri um conhecimento essencialmente teórico em várias áreas da indústria farmacêutica (financeira, desenvolvimento de negócio, assuntos regulamentares, investigação e desenvolvimento de medicamentos, garantia da qualidade, etc.) através da minha passagem pela Bluepharma. De seguida, na minha experiência de quatro meses na TREAT U, foi-me dada a oportunidade de realizar de forma mais independente, as funções inerentes ao cargo de assistente da gerência, com especial enfoque para atividades de gestão de projeto (incluindo assuntos regulamentares), tais como, apoio na preparação do plano de desenvolvimento não clínico e na preparação do pedido de aconselhamento científico. Este relatório começa assim por descrever os objectivos do estágio e uma breve descrição das instituições que me acolheram para a sua realização. De seguida, os conhecimentos adquiridos na vertente multidisciplinar do estágio e depois as atividades desenvolvidas no âmbito monodisciplinar. Por fim, apresenta uma análise das dificuldades e desafios encontrados bem como os esforços realizados para os ultrapassar.

Working with the complexities of adolescent mental health problems: applying Time‐limited Adolescent Psychodynamic Psychotherapy (TAPP)

This article discusses the application in a CAMHS setting of a distinctive intervention for adolescent mental health difficulties, Time‐limited Adolescent Psychodynamic Psychotherapy (TAPP). TAPP has been developed specifically for working with adolescents and the characteristic developmental and psychosocial complexities they present to mental health services. It is widely recognised that supporting the developmental process in adolescence is central to therapeutic interventions and the therapeutic aim of TAPP is to enable recovery of the capacity to meet developmental challenges. The key factors of TAPP are described, including the formulation and working with a developmental focus, the therapeutic stance, working with transference and counter‐transference, working with time limits, and the emphasis on engagement of adolescents in therapy in TAPP. The experiences of introducing and developing TAPP in the CAMHS service are discussed with two brief and one extended case examples and this leads to a discussion of the kinds of outcomes achieved. It is concluded that TAPP is a key and relevant intervention for adolescents in complex and vulnerable situations; further work will be undertaken to continue its application in these settings and to formally assess outcomes.

Conspicuous by their abstinence: the limited engagement of heroin users in English and Welsh Drug Recovery Wings

Background In recent years, an abstinence-focused, ‘recovery’ agenda has emerged in UK drug policy, largely in response to the perception that many opioid users had been ‘parked indefinitely’ on Opioid Substitution Therapy (OST). The introduction of ten pilot ‘Drug Recovery Wings’ (DRWs) in 2011 represents the application of this recovery agenda to prisons. This paper describes the DRWs’ operational models, the place of opiate dependent prisoners within them, and the challenges of delivering ‘recovery’ in prison. Methods In 2013, the implementation and operational models of all ten pilot DRWs were rapidly assessed. Up to three days were spent in each DRW, undertaking semi-structured interviews with a sample of 94 DRW staff and 102 DRW residents. Interviews were fully transcribed, and coded using grounded theory. Findings from the nine adult prisons are presented here. Results Four types of DRW were identified, distinguished by their size and selection criteria. Strikingly, no mid- or large-sized units regularly supported OST recipients through detoxification. Type A were large units whose residents were mostly on OST with long criminal records and few social or personal resources. Detoxification was rare, and medication reduction slow. Type B's mid-sized DRW was developed as a psychosocial support service for OST clients seeking detoxification. However, staff struggled to find such prisoners, and detoxification again proved rare. Type C DRWs focused on abstinence from all drugs, including OST. Though OST clients were not intentionally excluded, very few applied to these wings. Only Type D DRWs, offering intensive treatment on very small wings, regularly recruited OST recipients into abstinence-focused interventions. Conclusion Prison units wishing to support OST recipients in making greater progress towards abstinence may need to be small, intensive and take a stepped approach based on preparatory motivational work and extensive preparation for release. However, concerns about post-release deaths will remain.

Factoring in family freight forwarder company: case study

Each freight forwarder company must independently determine whether factoring is the right choice for them and if will pay off. If the factoring as a financial method would not be economically viable, it certainly would not exist. There are several different financing methods, which are more or less useful. However, not all of them are appropriate for every company, just like all buyers do not represent equal risks for the suppliers. The same buyer can pay one supplier within the agreed period, and not pay the other supplier at all. Still, the most important elements are the agreement made with the customer, the method of insuring ourselves and the method of monitoring the implementation of the agreements. Problems rarely occur overnight and that is why we have to constantly supervise our buyers; otherwise account receivables can occur. My thesis contributes to the ongoing research into how the use of factoring within the Slovenian economic environment can help a small freight forwarder company to grow. In effect, the specific characteristics of the Slovenian economic, legal and institutional environment make such an analysis an interesting one to explore. I used data from a family freight forwarder company TOP Hrvatin in drugi d.n.o. in order to establish future company growth by applying factoring as a right choice of financing. After applying all the data and preparing the company's financial plan, my results suggest that factoring is not the right choice for financing future investments for this particular freight forwarder business. Furthermore, the results even show that factoring is an extremely undeveloped financing method in Slovenia and that practically there is no legal base for its use.