Nonparametric Bayesian Density Modeling with Gaussian Processes

We present the Gaussian process density sampler (GPDS), an exchangeable generative model for use in nonparametric Bayesian density estimation. Samples drawn from the GPDS are consistent with exact, independent samples from a distribution defined by a density that is a transformation of a function drawn from a Gaussian process prior. Our formulation allows us to infer an unknown density from data using Markov chain Monte Carlo, which gives samples from the posterior distribution over density functions and from the predictive distribution on data space. We describe two such MCMC methods. Both methods also allow inference of the hyperparameters of the Gaussian process.

Fixed-lag blind equalization and sequence estimation in digital communications systems using sequential importance sampling

We present methods for fixed-lag smoothing using Sequential Importance sampling (SIS) on a discrete non-linear, non-Gaussian state space system with unknown parameters. Our particular application is in the field of digital communication systems. Each input data point is taken from a finite set of symbols. We represent transmission media as a fixed filter with a finite impulse response (FIR), hence a discrete state-space system is formed. Conventional Markov chain Monte Carlo (MCMC) techniques such as the Gibbs sampler are unsuitable for this task because they can only perform processing on a batch of data. Data arrives sequentially, so it would seem sensible to process it in this way. In addition, many communication systems are interactive, so there is a maximum level of latency that can be tolerated before a symbol is decoded. We will demonstrate this method by simulation and compare its performance to existing techniques.

Bayesian separation and recovery of convolutively mixed autoregressive sources

In this paper we address the problem of the separation and recovery of convolutively mixed autoregressive processes in a Bayesian framework. Solving this problem requires the ability to solve integration and/or optimization problems of complicated posterior distributions. We thus propose efficient stochastic algorithms based on Markov chain Monte Carlo (MCMC) methods. We present three algorithms. The first one is a classical Gibbs sampler that generates samples from the posterior distribution. The two other algorithms are stochastic optimization algorithms that allow to optimize either the marginal distribution of the sources, or the marginal distribution of the parameters of the sources and mixing filters, conditional upon the observation. Simulations are presented.

Bayesian model selection for time series using Markov chain Monte Carlo

We present a stochastic simulation technique for subset selection in time series models, based on the use of indicator variables with the Gibbs sampler within a hierarchical Bayesian framework. As an example, the method is applied to the selection of subset linear AR models, in which only significant lags are included. Joint sampling of the indicators and parameters is found to speed convergence. We discuss the possibility of model mixing where the model is not well determined by the data, and the extension of the approach to include non-linear model terms.

Bayesian learning of noisy Markov decision processes

This work addresses the problem of estimating the optimal value function in a Markov Decision Process from observed state-action pairs. We adopt a Bayesian approach to inference, which allows both the model to be estimated and predictions about actions to be made in a unified framework, providing a principled approach to mimicry of a controller on the basis of observed data. A new Markov chain Monte Carlo (MCMC) sampler is devised for simulation from theposterior distribution over the optimal value function. This step includes a parameter expansion step, which is shown to be essential for good convergence properties of the MCMC sampler. As an illustration, the method is applied to learning a human controller.

Cruise Report NOAA Ship McARTHUR II Cruise AR-04-04: Leg 2 (June 1-12, 2004): A pilot survey of deepwater coral/sponge assemblages and their susceptibility to fishing/harvest impacts at the Olympic Coast National Marine Sanctuary (OCNMS)

Summary: The offshore shelf and canyon habitats of the OCNMS (Fig. 1) are areas of high primary productivity and biodiversity that support extensive groundfish fisheries. Recent acoustic surveys conducted in these waters have indicated the presence of hard-bottom substrates believed to harbor unique deep-sea coral and sponge assemblages. Such fauna are often associated with shallow tropical waters, however an increasing number of studies around the world have recorded them in deeper, cold-water habitats in both northern and southern latitudes. These habitats are of tremendous value as sites of recruitment for commercially important fishes. Yet, ironically, studies have shown how the gear used in offshore demersal fishing, as well as other commercial operations on the seafloor, can cause severe physical disturbances to resident benthic fauna. Due to their exposed structure, slow growth and recruitment rates, and long life spans, deep-sea corals and sponges may be especially vulnerable to such disturbances, requiring very long periods to recover. Potential effects of fishing and other commercial operations in such critical habitats, and the need to define appropriate strategies for the protection of these resources, have been identified as a high-priority management issue for the sanctuary. To begin addressing this issue, an initial pilot survey was conducted June 1-12, 2004 at six sites in offshore waters of the OCNMS (Fig. 2, average depths of 147-265 m) to explore for the presence of deep-sea coral/sponge assemblages and to look for evidence of potential anthropogenic impacts in these critical habitats. The survey was conducted on the NOAA Ship McARTHUR-II using the Navy’s Phantom DHD2+2 remotely operated vehicle (ROV), which was equipped with a video camera, lasers, and a manipulator arm for the collection of voucher specimens. At each site, a 0.1-m2 grab sampler also was used to collect samples of sediments for the analysis of macroinfauna (> 1.0 mm), total organic carbon (TOC), grain size, and chemical contaminants. Vertical profiles of salinity, dissolved oxygen (DO), temperature, and pressure were recorded at each site with a small SeaCat conductivity-temperature-depth (CTD) profiler. Niskin bottles attached to the CTD also obtained near-bottom water samples in support of a companion study of microbial indicators of coral health and general ecological condition across these sites. All samples except the sediment-contaminant samples are being analyzed with present project funds. Original cruise plans included a total of 12 candidate stations to investigate (Fig. 3). However, inclement weather and equipment failures restricted the sampling to half of these sites. In spite of the limited sampling, the work completed was sufficient to address key project objectives and included several significant scientific observations. Foremost, the cruise was successful in demonstrating the presence of target deepwater coral species in these waters. Patches of the rare stony coral Lophelia pertusa, more characteristic of deepwater coral/sponge assemblages in the North Atlantic, were observed for the first time in OCNMS at a site in 271 meters of water. A large proportion of these corals consisted of dead and broken skeletal remains, and a broken gorgonian (soft coral) also was observed nearby. The source of these disturbances is not known. However, observations from several sites included evidence of bottom trawl marks in the sediment and derelict fishing gear (long lines). Preliminary results also support the view that these areas are important reservoirs of marine biodiversity and of value as habitat for demersal fishes. For example, onboard examination of 18 bottom-sediment grabs revealed benthic infaunal species representative of 14 different invertebrate phyla. Twenty-eight species of fishes from 11 families, including 11 (possibly 12) species of ommercially important rockfishes, also were identified from ROV video footage. These initial discoveries have sparked considerable interests in follow-up studies to learn more about the spatial extent of these assemblages and magnitude of potential impacts from commercial-fishing and other anthropogenic activities in the area. It is essential to expand our knowledge of these deep-sea communities and their vulnerability to potential environmental risks in order to determine the most appropriate management strategies. The survey was conducted under a partnership between NOAA’s National Centers for Coastal Ocean Science (NCCOS) and National Marine Sanctuary Program (NMSP) and included scientists from NCCOS, OCNMS, and several other west-coast State, academic, private, and tribal research institutions (see Section 4 for a complete listing of participating scientists). (PDF contains 20 pages)

Evaluation of salmon and steelhead spawning habitat quality in the South Fork Trinity River Basin, 1997

Sediment sampling was used to evaluate chinook salmon (Oncorhynchus tshawytscha) and steelhead (O. mykiss) spawning habitat quality in the South Fork Trinity River (SFTR) basin. Sediment samples were collected using a McNeil-type sampler and wet sieved through a series of Tyler screens (25.00 mm, 12.50 mm, 6.30 mm, 3.35 mm, 1.00 mm, and 0.85 mm). Fines (particles < 0.85 mm) were determined after a l0-minute settling period in Imhoff cones. Thirteen stations were sampled in the SFTR basin: five stations were located in mainstem SFTR between rk 2.1 and 118.5, 2 stations each were located in EF of the SFTR, Grouse Creek, and Madden Creek, and one station each was located in Eltapom and Hayfork Creeks. Sample means for fines(particles < 0.85 mm) fer SFTR stations ranged between 14.4 and 19.4%; tributary station sample mean fines ranged between 3.4 and 19.4%. Decreased egg survival would be expected at 4 of 5 mainstem SFTR stations and at one station in EF of SFTR and Grouse Creek where fines content exceed 15%. Small gravel/sand content measured at all stations were high, and exceed levels associated with reduced sac fry emergence rates. Reduction of egg survival or sac fry emergence due to sedimentation in spawning gravels could lead to reduced juvenile production from the South Fork Trinity River. (PDF contains 18 pages.)

The power of quantum Fourier sampling

How powerful are Quantum Computers? Despite the prevailing belief that Quantum Computers are more powerful than their classical counterparts, this remains a conjecture backed by little formal evidence. Shor's famous factoring algorithm [Shor97] gives an example of a problem that can be solved efficiently on a quantum computer with no known efficient classical algorithm. Factoring, however, is unlikely to be NP-Hard, meaning that few unexpected formal consequences would arise, should such a classical algorithm be discovered. Could it then be the case that any quantum algorithm can be simulated efficiently classically? Likewise, could it be the case that Quantum Computers can quickly solve problems much harder than factoring? If so, where does this power come from, and what classical computational resources do we need to solve the hardest problems for which there exist efficient quantum algorithms?

We make progress toward understanding these questions through studying the relationship between classical nondeterminism and quantum computing. In particular, is there a problem that can be solved efficiently on a Quantum Computer that cannot be efficiently solved using nondeterminism? In this thesis we address this problem from the perspective of sampling problems. Namely, we give evidence that approximately sampling the Quantum Fourier Transform of an efficiently computable function, while easy quantumly, is hard for any classical machine in the Polynomial Time Hierarchy. In particular, we prove the existence of a class of distributions that can be sampled efficiently by a Quantum Computer, that likely cannot be approximately sampled in randomized polynomial time with an oracle for the Polynomial Time Hierarchy.

Our work complements and generalizes the evidence given in Aaronson and Arkhipov's work [AA2013] where a different distribution with the same computational properties was given. Our result is more general than theirs, but requires a more powerful quantum sampler.

Modelling soil erosion and transport in the Burrishoole catchment, Newport, Co. Mayo, Ireland

The Burrishoole catchment is situated in County Mayo, on the northwest coast of the Republic of Ireland. Much of the catchment is covered by blanket peat that, in many areas, has become heavily eroded in recent years. This is thought to be due, primarily, to the adverse effects of forestry and agricultural activities in the area. Such activities include ploughing, drainage, the planting and harvesting of trees, and sheep farming, all of which are potentially damaging to such a sensitive landscape if not managed carefully. This article examines the sediment yield and hydrology of the Burrishoole catchment. Flow and sediment concentrations were measured at 8-hourly intervals from 5 February 2001 to 8 November 2001 with an automatic sampler and separate flow gauge, and hourly averages were recorded between 4 July 2002 and 6 September 2002 using an automatic river monitoring system [ARMS]. The authors describe the GIS-based model of soil erosion and transport that was applied to the Burrishoole catchment during this study. The results of these analyses were compared, in a qualitative manner, with the aerial photography available for the Burrishoole catchment to see whether areas that were predicted to contribute large proportions of eroded material to the drainage network corresponded with areas where peat erosion could be identified through photo-interpretation.

Exposição ao material particulado 2,5 m coletado em vias de alto tráfego da cidade do Rio de Janeiro: avaliação mutagênica, genotóxica e determinação de risco à saúde induzido por hidrocarbonetos policíclicos aromáticos

A mutagenicidade do material particulado é atribuída primeiramente aos hidrocarbonetos policíclicos aromáticos (HPA). Investigamos a atividade mutagênica do material particulado (MP2,5) em amostras coletadas em três pontos da cidade do Rio de Janeiro. As coletas foram realizadas com auxílio de um amostrador de grande volume na Avenida Brasil, no campus da Universidade do Estado do Rio de Janeiro e no Túnel Rebouças em filtros de fibra de vidro. Metade de cada filtro foi submetido à extração por sonicação com o solvente diclorometano. Seis HPA foram identificados e quantificados por cromatografia gasosa com espectrometria de massa (GC/MS). Após a análise química as concentrações dos HPA obtidos foram correlacionados ao fatores físicos, além de ser realizado avaliação de risco para cada HPA estudado. Linhagens de Salmonella typhimurium (TA98 e derivadas TA98/1.8-DNP6, YG1021 e YG1024) foram utilizadas no ensaio de mutagenicidade e tratadas (10-50 g/placa) com extrato orgânico na presença e na ausência de metabolização exógena. Células de raiz de cebola foram tratadas com extratos orgânicos nas concentrações (5-25g/mL). A alta umidade encontrada no Túnel Rebouças pode ter influenciado na deposição de cinco dos seis HPA estudados em material particulado. Além disso, em diferentes condições de tráfego, motoristas de ônibus que cruzam a Avenida Brasil e o Rebouças túnel estão expostos ao risco induzidos por HPA na ordem de 10-6. Mutagenicidade foi detectada tanto na presença quanto na ausência de metabolização, para as linhagens YG1021 e YG1024 nos três pontos, sugerindo a presença de nitro e amino derivados de HPA. As amostras do Túnel Rebouças apresentaram os maiores valores para rev/g e rev/m3. Estes resultados podem estar relacionados ao longo trajeto e a restrita ventilação. Efeito citotóxico foi detectado pelo ensaio Allium cepa nos três pontos de monitoramento. Além disso os extratos orgânicos provenientes das coletas da Avenida Brasil, UERJ e do Túnel Rebouças induziram efeito clastogênico em células de raiz de Allium cepa

Étude comparative de la macrofaune associée aux racines de Pistia stratiotes L. (Araceae) du lac de barrage d'Ayamé II et du fleuve Comoé (Côte d'Ivoire)

Abundance and species variations of the macrofauna associated with the roots of Pistia stratiotes of Ayamé II lake and Comoé river were studied during one year. Samples of plants were collected by hand once a month within a sampler of 1 m2 of surface.

Temporal Changes in a Tropical Nekton Assemblage and Performance of a Prawn Selective Gear

The temporal variation of components of a moderately diverse (H=1.46) tropical estuarine fish assemblage (long. 146°30'E, lat. 8°45'S) was directed by salinities that had been determined by local oceanographic and probably topographic conditions. For this assemblage, two types of intrayear component profiles are predicted. Pooled data (1988-91) reveal a large component of regular/resident species (43%) in an assemblage which has been under a narrow temperature regime «5T). These results facilitate a discussion on the relevance and usefulness of three hypotheses often cited in studies concerning species diversity and component characteristics of the subtropical/tropical coastal nonreef fish assemblages. Manifestations of the assemblage are reflected in catch composition and weights of 39 trials conducted for a selective prawning gear whose performance in bycatch reduction, mainly for finfishes, is judged by an index, E, we have previously proposed. This gear is capable of harvesting the prawn while conserving the demersal fish. Behavioral responses to netting of the prawns and the finfishes, especially the nearshore surface schoolers such as leiognathids, are discussed from several points of view. An adaptation in terms of group selection for leiognathids of their locking mechanism of median fin spines has been interpreted. For the purpose of bycatch reduction or E enhancement, suggestions for improvements in net design and trawl configuration by considering the behavioral features of fish are made. Our original formula of E is modified for general use. Bycatch problems in the regional prawn fisheries and their possible impacts on fishery planning and development in Papua New Guinea as a developing country are discussed. The gear tested may offer enormous ecological and economic benefits. The gear is multipurpose, extremely simple, and can also be used as a biological sampler.

Sediment flux at selected coastal sites: proposed time series measurment by particle traps

Particle flux in the ocean reflects ongoing biological and geological processes operating under the influence of the local environment. Estimation of this particle flux through sediment trap deployment is constrained by sampler accuracy, particle preservation, and swimmer distortion. Interpretation of specific particle flux is further constrained by indeterminate particle dispersion and the absence of a clear understanding of the sedimentary consequences of ecosystem activity. Nevertheless, the continuous and integrative properties of the particle trap measure, along with the logistic advantage of a long-term moored sampler, provide a set of strategic advantages that appear analogous to those underlying conventional oceanographic survey programs. Emboldened by this perception, several stations along the coast of Southern California and Mexico have been targeted as coastal ocean flux sites (COFS).

Pitman-Yor Diffusion Trees

We introduce the Pitman Yor Diffusion Tree (PYDT) for hierarchical clustering, a generalization of the Dirichlet Diffusion Tree (Neal, 2001) which removes the restriction to binary branching structure. The generative process is described and shown to result in an exchangeable distribution over data points. We prove some theoretical properties of the model and then present two inference methods: a collapsed MCMC sampler which allows us to model uncertainty over tree structures, and a computationally efficient greedy Bayesian EM search algorithm. Both algorithms use message passing on the tree structure. The utility of the model and algorithms is demonstrated on synthetic and real world data, both continuous and binary.

Bayesian approach to signal modelling

In this paper, an introduction to Bayesian methods in signal processing will be given. The paper starts by considering the important issues of model selection and parameter estimation and derives analytic expressions for the model probabilities of two simple models. The idea of marginal estimation of certain model parameter is then introduced and expressions are derived for the marginal probability densities for frequencies in white Gaussian noise and a Bayesian approach to general changepoint analysis is given. Numerical integration methods are introduced based on Markov chain Monte Carlo techniques and the Gibbs sampler in particular.