Learning dependent sources using mixtures of Dirichlet: applications on hyperspectral unmixing

This paper is an elaboration of the DECA algorithm [1] to blindly unmix hyperspectral data. The underlying mixing model is linear, meaning that each pixel is a linear mixture of the endmembers signatures weighted by the correspondent abundance fractions. The proposed method, as DECA, is tailored to highly mixed mixtures in which the geometric based approaches fail to identify the simplex of minimum volume enclosing the observed spectral vectors. We resort then to a statitistical framework, where the abundance fractions are modeled as mixtures of Dirichlet densities, thus enforcing the constraints on abundance fractions imposed by the acquisition process, namely non-negativity and constant sum. With respect to DECA, we introduce two improvements: 1) the number of Dirichlet modes are inferred based on the minimum description length (MDL) principle; 2) The generalized expectation maximization (GEM) algorithm we adopt to infer the model parameters is improved by using alternating minimization and augmented Lagrangian methods to compute the mixing matrix. The effectiveness of the proposed algorithm is illustrated with simulated and read data.

Scheduling of head-dependent cascaded hydro systems: Mixed-integer quadratic programming approach

This paper is on the problem of short-term hydro, scheduling, particularly concerning head-dependent cascaded hydro systems. We propose a novel mixed-integer quadratic programming approach, considering not only head-dependency, but also discontinuous operating regions and discharge ramping constraints. Thus, an enhanced short-term hydro scheduling is provided due to the more realistic modeling presented in this paper. Numerical results from two case studies, based on Portuguese cascaded hydro systems, illustrate the proficiency of the proposed approach.

Bias dependent photocurrent collection in p-i-n a-Si : EVSiC : H heterojunction

A series of large area single layers and heterojunction cells in the assembly glass/ZnO:Al/p (SixC1-x:H)/i (Si:H)/n (SixC1-x:H)/Al (0

Bias-dependent photocurrent collection in p-i-n a-Si : H/SiC : H heterojunction

A series of large area single layers and glass/ZnO:AVp(SixC1-x:H)/i(Si:H)/n(SixC1-x:H)/AI (0 < x < 1) heterojunction cells were produced by plasma-enhanced chemical vapour deposition (PE-CVD) at low temperature. Junction properties, carrier transport and photogeneration are investigated from dark and illuminated current-voltage (J-V) and capacitance-voltage (C-V) characteristics. For the heterojunction cells atypical J-V characteristics under different illumination conditions are observed leading to poor fill factors. High series resistances around 106 Q are also measured. These experimental results were used as a basis for the numerical simulation of the energy band diagram, and the electrical field distribution of the structures. Further comparison with the sensor performance gave satisfactory agreement. Results show that the conduction band offset is the most limiting parameter for the optimal collection of the photogenerated carriers. As the optical gap increases and the conductivity of the doped layers decreases, the transport mechanism changes from a drift to a diffusion-limited process.

Malaria transmission model for different levels of acquired immunity and temperature-dependent parameters (vector)

OBJECTIVE: Describe the overall transmission of malaria through a compartmental model, considering the human host and mosquito vector. METHODS: A mathematical model was developed based on the following parameters: human host immunity, assuming the existence of acquired immunity and immunological memory, which boosts the protective response upon reinfection; mosquito vector, taking into account that the average period of development from egg to adult mosquito and the extrinsic incubation period of parasites (transformation of infected but non-infectious mosquitoes into infectious mosquitoes) are dependent on the ambient temperature. RESULTS: The steady state equilibrium values obtained with the model allowed the calculation of the basic reproduction ratio in terms of the model's parameters. CONCLUSIONS: The model allowed the calculation of the basic reproduction ratio, one of the most important epidemiological variables.

Re-entrant phase behaviour of network fluids: A patchy particle model with temperature-dependent valence

We study a model consisting of particles with dissimilar bonding sites ("patches"), which exhibits self-assembly into chains connected by Y-junctions, and investigate its phase behaviour by both simulations and theory. We show that, as the energy cost epsilon(j) of forming Y-junctions increases, the extent of the liquid-vapour coexistence region at lower temperatures and densities is reduced. The phase diagram thus acquires a characteristic "pinched" shape in which the liquid branch density decreases as the temperature is lowered. To our knowledge, this is the first model in which the predicted topological phase transition between a fluid composed of short chains and a fluid rich in Y-junctions is actually observed. Above a certain threshold for epsilon(j), condensation ceases to exist because the entropy gain of forming Y-junctions can no longer offset their energy cost. We also show that the properties of these phase diagrams can be understood in terms of a temperature-dependent effective valence of the patchy particles. (C) 2011 American Institute of Physics. [doi: 10.1063/1.3605703]

Cost dependent strategy for electricity markets bidding based on adaptive reinforcement learning

Electricity markets are complex environments, involving a large number of different entities, playing in a dynamic scene to obtain the best advantages and profits. MASCEM is a multi-agent electricity market simulator to model market players and simulate their operation in the market. Market players are entities with specific characteristics and objectives, making their decisions and interacting with other players. MASCEM provides several dynamic strategies for agents’ behavior. This paper presents a method that aims to provide market players with strategic bidding capabilities, allowing them to obtain the higher possible gains out of the market. This method uses a reinforcement learning algorithm to learn from experience how to choose the best from a set of possible bids. These bids are defined accordingly to the cost function that each producer presents.

Dependent component analysis: a hyperspectral unmixing algorithm

Linear unmixing decomposes a hyperspectral image into a collection of reflectance spectra of the materials present in the scene, called endmember signatures, and the corresponding abundance fractions at each pixel in a spatial area of interest. This paper introduces a new unmixing method, called Dependent Component Analysis (DECA), which overcomes the limitations of unmixing methods based on Independent Component Analysis (ICA) and on geometrical properties of hyperspectral data. DECA models the abundance fractions as mixtures of Dirichlet densities, thus enforcing the constraints on abundance fractions imposed by the acquisition process, namely non-negativity and constant sum. The mixing matrix is inferred by a generalized expectation-maximization (GEM) type algorithm. The performance of the method is illustrated using simulated and real data.

Optimising and adapting the QoS of a dynamic set of inter-dependent tasks

Due to the growing complexity and adaptability requirements of real-time systems, which often exhibit unrestricted Quality of Service (QoS) inter-dependencies among supported services and user-imposed quality constraints, it is increasingly difficult to optimise the level of service of a dynamic task set within an useful and bounded time. This is even more difficult when intending to benefit from the full potential of an open distributed cooperating environment, where service characteristics are not known beforehand and tasks may be inter-dependent. This paper focuses on optimising a dynamic local set of inter-dependent tasks that can be executed at varying levels of QoS to achieve an efficient resource usage that is constantly adapted to the specific constraints of devices and users, nature of executing tasks and dynamically changing system conditions. Extensive simulations demonstrate that the proposed anytime algorithms are able to quickly find a good initial solution and effectively optimise the rate at which the quality of the current solution improves as the algorithms are given more time to run, with a minimum overhead when compared against their traditional versions.

Dynamic QoS adaptation of inter- dependent task sets in cooperative embedded systems

Due to the growing complexity and dynamism of many embedded application domains (including consumer electronics, robotics, automotive and telecommunications), it is increasingly difficult to react to load variations and adapt the system's performance in a controlled fashion within an useful and bounded time. This is particularly noticeable when intending to benefit from the full potential of an open distributed cooperating environment, where service characteristics are not known beforehand and tasks may exhibit unrestricted QoS inter-dependencies. This paper proposes a novel anytime adaptive QoS control policy in which the online search for the best set of QoS levels is combined with each user's personal preferences on their services' adaptation behaviour. Extensive simulations demonstrate that the proposed anytime algorithms are able to quickly find a good initial solution and effectively optimise the rate at which the quality of the current solution improves as the algorithms are given more time to run, with a minimum overhead when compared against their traditional versions.

Inhibition of Aspergillus fumigatus and its biofilm by Pseudomonas aeruginosa is dependent on the source, phenotype and growth conditions of the bacterium

Aspergillus fumigatus (Af) and Pseudomonas aeruginosa (Pa) are leading fungal and bacterial pathogens, respectively, in many clinical situations. Relevant to this, their interface and co-existence has been studied. In some experiments in vitro, Pa products have been defined that are inhibitory to Af. In some clinical situations, both can be biofilm producers, and biofilm could alter their physiology and affect their interaction. That may be most relevant to airways in cystic fibrosis (CF), where both are often prominent residents. We have studied clinical Pa isolates from several sources for their effects on Af, including testing involving their biofilms. We show that the described inhibition of Af is related to the source and phenotype of the Pa isolate. Pa cells inhibited the growth and formation of Af biofilm from conidia, with CF isolates more inhibitory than non-CF isolates, and non-mucoid CF isolates most inhibitory. Inhibition did not require live Pa contact, as culture filtrates were also inhibitory, and again non-mucoid>mucoid CF>non-CF. Preformed Af biofilm was more resistant to Pa, and inhibition that occurred could be reproduced with filtrates. Inhibition of Af biofilm appears also dependent on bacterial growth conditions; filtrates from Pa grown as biofilm were more inhibitory than from Pa grown planktonically. The differences in Pa shown from these different sources are consistent with the extensive evolutionary Pa changes that have been described in association with chronic residence in CF airways, and may reflect adaptive changes to life in a polymicrobial environment.

Brownian bridge and other path-dependent Gaussian processes vectorial simulation

The iterative simulation of the Brownian bridge is well known. In this article, we present a vectorial simulation alternative based on Gaussian processes for machine learning regression that is suitable for interpreted programming languages implementations. We extend the vectorial simulation of path-dependent trajectories to other Gaussian processes, namely, sequences of Brownian bridges, geometric Brownian motion, fractional Brownian motion, and Ornstein-Ulenbeck mean reversion process.

Estrogen receptor dependent genetic and epigenetic factors of tamoxifen resistance

Resumo: A decisão da terapêutica hormonal no tratamento do cancro da mama baseiase na determinação do receptor de estrogénio alfa por imunohistoquímica (IHC). Contudo, a presença deste receptor não prediz a resposta em todas as situações, em parte devido a limitações do método IHC. Investigámos se a expressão dos genes ESR1 e ESR2, bem como a metilação dos respectivos promotores, pode estar relacionada com a evolução desfavorável de uma proporção de doentes tratados com tamoxifeno assim como com a perda dos receptores de estrogénio alfa (ERα) e beta (ERß). Amostras de 211 doentes com cancro da mama diagnosticado entre 1988 e 2004, fixadas em formalina e preservadas em parafina, foram utilizadas para a determinação por IHC da presença dos receptores ERα e ERß. O mRNA total do gene ESR1 e os níveis específicos do transcrito derivado do promotor C (ESR1_C), bem como dos transcritos ESR2_ß1, ESR2_ß2/cx, and ESR2_ß5 foram avaliados por Real-time PCR. Os promotores A e C do gene ESR1 e os promotores 0K e 0N do gene ESR2 foram investigados por análise de metilação dos dinucleotidos CpG usando bisulfite-PCR para análise com enzimas de restrição, ou para methylation specific PCR. Atendendo aos resultados promissores relacionados com a metilação do promotor do gene ESR1, complementamos o estudo com um método quantitativo por matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) suportado pelo software Epityper para a medição da metilação nos promotores A e C. Fez-se a avaliação da estabilidade do mRNA nas linhas celulares de cancro da mama MCF-7 e MDA-MB-231 tratadas com actinomicina D. Baixos níveis do transcrito ESR1_C associaram-se a uma melhor sobrevivência global (p = 0.017). Níveis elevados do transcrito ESR1_C associaram-se a uma resposta inferior ao tamoxifeno (HR = 2.48; CI 95% 1.24-4.99), um efeito mais pronunciado em doentes com tumores de fenótipo ERα/PgR duplamente positivo (HR = 3.41; CI 95% 1.45-8.04). A isoforma ESR1_C mostrou ter uma semi-vida prolongada, bem como uma estrutura secundária da região 5’UTR muito mais relaxada em comparação com a isoforma ESR1_A. A análise por Western-blot mostrou que ao nível da 21 proteína, a selectividade de promotores é indistinguivel. Não se detectou qualquer correlação entre os níveis das isoformas do gene ESR2 ou entre a metilação dos promotores do gene ESR2, e a detecção da proteína ERß. A metilação do promotor C do gene ESR1, e não do promotor A, foi responsável pela perda do receptor ERα. Estes resultados sugerem que os níveis do transcrito ESR1_C sejam usados como um novo potencial marcador para o prognóstico e predição de resposta ao tratamento com tamoxifeno em doentes com cancro da mama. Abstract: The decision of endocrine breast cancer treatment relies on ERα IHC-based assessment. However, ER positivity does not predict response in all cases in part due to IHC methodological limitations. We investigated whether ESR1 and ESR2 gene expression and respective promoter methylation may be related to non-favorable outcome of a proportion of tamoxifen treated patients as well as to ERα and ERß loss. Formalin-fixed paraffin-embedded breast cancer samples from 211 patients diagnosed between 1988 and 2004 were submitted to IHC-based ERα and ERß protein determination. ESR1 whole mRNA and promoter C specific transcript levels, as well as ESR2_ß1, ESR2_ß2/cx, and ESR2_ß5 transcripts were assessed by real-time PCR. ESR1 promoters A and C, and ESR2 promoters 0N and 0K were investigated by CpG methylation analysis using bisulfite-PCR for restriction analysis, or methylation specific PCR. Due to the promising results related to ESR1 promoter methylation, we have used a quantification method by matrixassisted laser desorption/ionization time-of-flight mass spectrometry (MALDITOF MS) together with Epityper software to measure methylation at promoters A and C. mRNA stability was assessed in actinomycin D treated MCF-7 and MDA-MB-231 cells. ERα protein was quantified using transiently transfected breast cancer cells. Low ESR1_C transcript levels were associated with better overall survival (p = 0.017). High levels of ESR1_C transcript were associated with non-favorable response in tamoxifen treated patients (HR = 2.48; CI 95% 1.24-4.99), an effect that was more pronounced in patients with ERα/PgR double-positive tumors (HR = 3.41; CI 95% 1.45-8.04). The ESR1_C isoform had a prolonged mRNA half-life and a more relaxed 5’UTR structure compared to ESR1_A isoform. Western-blot analysis showed that at protein level, the promoter selectivity is undistinguishable. There was no correlation between levels of ESR2 isoforms or ESR2 promoter methylation and ERß protein staining. ESR1 promoter C CpG methylation and not promoter A was responsible for ERα loss. We propose ESR1_C levels as a putative novel marker for breast cancer prognosis and prediction of tamoxifen response.

Unmixing hyperspectral data: independent and dependent component analysis

The development of high spatial resolution airborne and spaceborne sensors has improved the capability of ground-based data collection in the fields of agriculture, geography, geology, mineral identification, detection [2, 3], and classification [4–8]. The signal read by the sensor from a given spatial element of resolution and at a given spectral band is a mixing of components originated by the constituent substances, termed endmembers, located at that element of resolution. This chapter addresses hyperspectral unmixing, which is the decomposition of the pixel spectra into a collection of constituent spectra, or spectral signatures, and their corresponding fractional abundances indicating the proportion of each endmember present in the pixel [9, 10]. Depending on the mixing scales at each pixel, the observed mixture is either linear or nonlinear [11, 12]. The linear mixing model holds when the mixing scale is macroscopic [13]. The nonlinear model holds when the mixing scale is microscopic (i.e., intimate mixtures) [14, 15]. The linear model assumes negligible interaction among distinct endmembers [16, 17]. The nonlinear model assumes that incident solar radiation is scattered by the scene through multiple bounces involving several endmembers [18]. Under the linear mixing model and assuming that the number of endmembers and their spectral signatures are known, hyperspectral unmixing is a linear problem, which can be addressed, for example, under the maximum likelihood setup [19], the constrained least-squares approach [20], the spectral signature matching [21], the spectral angle mapper [22], and the subspace projection methods [20, 23, 24]. Orthogonal subspace projection [23] reduces the data dimensionality, suppresses undesired spectral signatures, and detects the presence of a spectral signature of interest. The basic concept is to project each pixel onto a subspace that is orthogonal to the undesired signatures. As shown in Settle [19], the orthogonal subspace projection technique is equivalent to the maximum likelihood estimator. This projection technique was extended by three unconstrained least-squares approaches [24] (signature space orthogonal projection, oblique subspace projection, target signature space orthogonal projection). Other works using maximum a posteriori probability (MAP) framework [25] and projection pursuit [26, 27] have also been applied to hyperspectral data. In most cases the number of endmembers and their signatures are not known. Independent component analysis (ICA) is an unsupervised source separation process that has been applied with success to blind source separation, to feature extraction, and to unsupervised recognition [28, 29]. ICA consists in finding a linear decomposition of observed data yielding statistically independent components. Given that hyperspectral data are, in given circumstances, linear mixtures, ICA comes to mind as a possible tool to unmix this class of data. In fact, the application of ICA to hyperspectral data has been proposed in reference 30, where endmember signatures are treated as sources and the mixing matrix is composed by the abundance fractions, and in references 9, 25, and 31–38, where sources are the abundance fractions of each endmember. In the first approach, we face two problems: (1) The number of samples are limited to the number of channels and (2) the process of pixel selection, playing the role of mixed sources, is not straightforward. In the second approach, ICA is based on the assumption of mutually independent sources, which is not the case of hyperspectral data, since the sum of the abundance fractions is constant, implying dependence among abundances. This dependence compromises ICA applicability to hyperspectral images. In addition, hyperspectral data are immersed in noise, which degrades the ICA performance. IFA [39] was introduced as a method for recovering independent hidden sources from their observed noisy mixtures. IFA implements two steps. First, source densities and noise covariance are estimated from the observed data by maximum likelihood. Second, sources are reconstructed by an optimal nonlinear estimator. Although IFA is a well-suited technique to unmix independent sources under noisy observations, the dependence among abundance fractions in hyperspectral imagery compromises, as in the ICA case, the IFA performance. Considering the linear mixing model, hyperspectral observations are in a simplex whose vertices correspond to the endmembers. Several approaches [40–43] have exploited this geometric feature of hyperspectral mixtures [42]. Minimum volume transform (MVT) algorithm [43] determines the simplex of minimum volume containing the data. The MVT-type approaches are complex from the computational point of view. Usually, these algorithms first find the convex hull defined by the observed data and then fit a minimum volume simplex to it. Aiming at a lower computational complexity, some algorithms such as the vertex component analysis (VCA) [44], the pixel purity index (PPI) [42], and the N-FINDR [45] still find the minimum volume simplex containing the data cloud, but they assume the presence in the data of at least one pure pixel of each endmember. This is a strong requisite that may not hold in some data sets. In any case, these algorithms find the set of most pure pixels in the data. Hyperspectral sensors collects spatial images over many narrow contiguous bands, yielding large amounts of data. For this reason, very often, the processing of hyperspectral data, included unmixing, is preceded by a dimensionality reduction step to reduce computational complexity and to improve the signal-to-noise ratio (SNR). Principal component analysis (PCA) [46], maximum noise fraction (MNF) [47], and singular value decomposition (SVD) [48] are three well-known projection techniques widely used in remote sensing in general and in unmixing in particular. The newly introduced method [49] exploits the structure of hyperspectral mixtures, namely the fact that spectral vectors are nonnegative. The computational complexity associated with these techniques is an obstacle to real-time implementations. To overcome this problem, band selection [50] and non-statistical [51] algorithms have been introduced. This chapter addresses hyperspectral data source dependence and its impact on ICA and IFA performances. The study consider simulated and real data and is based on mutual information minimization. Hyperspectral observations are described by a generative model. This model takes into account the degradation mechanisms normally found in hyperspectral applications—namely, signature variability [52–54], abundance constraints, topography modulation, and system noise. The computation of mutual information is based on fitting mixtures of Gaussians (MOG) to data. The MOG parameters (number of components, means, covariances, and weights) are inferred using the minimum description length (MDL) based algorithm [55]. We study the behavior of the mutual information as a function of the unmixing matrix. The conclusion is that the unmixing matrix minimizing the mutual information might be very far from the true one. Nevertheless, some abundance fractions might be well separated, mainly in the presence of strong signature variability, a large number of endmembers, and high SNR. We end this chapter by sketching a new methodology to blindly unmix hyperspectral data, where abundance fractions are modeled as a mixture of Dirichlet sources. This model enforces positivity and constant sum sources (full additivity) constraints. The mixing matrix is inferred by an expectation-maximization (EM)-type algorithm. This approach is in the vein of references 39 and 56, replacing independent sources represented by MOG with mixture of Dirichlet sources. Compared with the geometric-based approaches, the advantage of this model is that there is no need to have pure pixels in the observations. The chapter is organized as follows. Section 6.2 presents a spectral radiance model and formulates the spectral unmixing as a linear problem accounting for abundance constraints, signature variability, topography modulation, and system noise. Section 6.3 presents a brief resume of ICA and IFA algorithms. Section 6.4 illustrates the performance of IFA and of some well-known ICA algorithms with experimental data. Section 6.5 studies the ICA and IFA limitations in unmixing hyperspectral data. Section 6.6 presents results of ICA based on real data. Section 6.7 describes the new blind unmixing scheme and some illustrative examples. Section 6.8 concludes with some remarks.

Hyperspectral unmixing algorithm via dependent component analysis

This paper introduces a new method to blindly unmix hyperspectral data, termed dependent component analysis (DECA). This method decomposes a hyperspectral images into a collection of reflectance (or radiance) spectra of the materials present in the scene (endmember signatures) and the corresponding abundance fractions at each pixel. DECA assumes that each pixel is a linear mixture of the endmembers signatures weighted by the correspondent abundance fractions. These abudances are modeled as mixtures of Dirichlet densities, thus enforcing the constraints on abundance fractions imposed by the acquisition process, namely non-negativity and constant sum. The mixing matrix is inferred by a generalized expectation-maximization (GEM) type algorithm. This method overcomes the limitations of unmixing methods based on Independent Component Analysis (ICA) and on geometrical based approaches. The effectiveness of the proposed method is illustrated using simulated data based on U.S.G.S. laboratory spectra and real hyperspectral data collected by the AVIRIS sensor over Cuprite, Nevada.