990 resultados para Gaussian channel
Resumo:
Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets. © 2010 Springer-Verlag.
Resumo:
The present paper investigates the occurrence and abundance of brachyuran larvae in the Manora Channel during August- December 1993. The fortnightly planktonic sampling was carried out in Manora Channel which is bordered by mangroves, during day time from surface and subsurface waters at shallow depth using Bongo net of 300 micron mesh size. Analysis of samples revealed presence of brachyuran larvae of 12 species belonging to 6 families. Of these 4 species are confirmed: Serenella indica, Dotilla blanfordi, Metopograpsus thukuhar and Clistocoeloma lanatum, 2 provisionally identified species are: Pilumnus ?karachiensis and Pinnotheres ?pisum, 2 species are identified upto generic level: Philyra sp. and Pinnotheres sp., and 1 Ocypodid species and 3 Xanthid species are identified upto family level. This study based on identification, occurrence and abundance of brachyuran larvae in the area, also gives percentage composition of brachyuran larvae collected during 1993, in the Manora Channel.
Resumo:
Occurrence and abundance of Pseudo-nitzschia spp. at the mouth of the Maheshkhali channel of the Bay of Bengal, Bangladesh were studied. Plankton and water samples were collected monthly from the sub-surface water during high tide at daytime from June 2000 to May 2001. Four species of Pseudo-nitzschia, namely P. pungens, P. pseudodelicatissima, P. delicatissima and P. australis were identified and among them the first three were most commonly encountered and they varied seasonally. Pseudo-nitzschia delicatissima was the dominant species during the autumn and winter months, whereas P. pungens was dominant during the summer and spring months. Pseudo-nitzschia pseudodelicatissima exhibited its highest abundance level during the summer. Surface water temperature, salinity, nitrate-nitrogen (N03-N) and phosphate-phosphorus (P04-P) were recorded and their relationship with the occurrence and abundance of Pseudo-nitzschia species were studied. At the mouth of the Maheshkhali channel, Pseudo-nitzschia cell density was highest in late autumn (November) when highest salinity (35 o/oo) and P04-P (3.2 mg/l) concentrations and low temperature (23 °C) were recorded.
Resumo:
The seasonal mean size distribution of A. chinensis were estimated as 29.229mm ±4.77, 25.125mm ±2.55, 25.165mm ±2.29 and 32.44mm ±3.63 for annual, monsoon, postmonsoon and pre-monsoon period, respectively. Seasonal mean carapace length distribution were estimated as 9.37mm ±1.457, 8.063mm ±0.63,8.258mm ±0.59 and 10.37mm ±l.ll3 for annual, monsoon, post-monsoon and the pre-monsoon season. The carapace length and total length relationships was found to be TL= - 1.39±3.23 CL. Linear relation was found in arithmetic and as well as logarithmic scale.
Resumo:
In total 68 phytoplankton species were identified at the mouth of the Maheshkhali channel with the Bay of Bengal, among them 41 belong to Bacillariophyceae, 17 Dinophyceae, 7 Cyanophyceae and 3 to Chlorophyceae. The highest phytoplankton production was observed in November (578.0 x 105 cells/L) and the lowest in June (37.5 x 105 cells/L). Some hydrographic parameters e.g., surface water temperature, salinity and nutrients (N03-N and P04-P) were recorded and their relationship with the occurrence and abundance of phytoplankton population were also studied. Nutrient concentration was higher during the autumn months, when rain water provided the maximum outflow of rivers discharging into the channel. During the nutrient peak period, the total phytoplankton production was maximum. Bacillariophyceae was the dominant group of phytoplankton throughout the study period except in June and September, when Dinophyceae was dominant. Cyanophyceae was abundant in spring months when temperature began to rise.
Resumo:
We introduce a new regression framework, Gaussian process regression networks (GPRN), which combines the structural properties of Bayesian neural networks with the non-parametric flexibility of Gaussian processes. This model accommodates input dependent signal and noise correlations between multiple response variables, input dependent length-scales and amplitudes, and heavy-tailed predictive distributions. We derive both efficient Markov chain Monte Carlo and variational Bayes inference procedures for this model. We apply GPRN as a multiple output regression and multivariate volatility model, demonstrating substantially improved performance over eight popular multiple output (multi-task) Gaussian process models and three multivariate volatility models on benchmark datasets, including a 1000 dimensional gene expression dataset.
Resumo:
We report high hole and electron mobilities in nanocrystalline silicon (nc-Si:H) top-gate staggered thin-film transistors (TFTs) fabricated by direct plasma-enhanced chemical vapor deposition (PECVD) at 260°C. The n-channel nc-Si:H TFT with n+ nc-Si:H ohmic contacts shows a field-effect electron mobility (μnFE) of 130 cm2/Vs, which increases to 150 cm2/Vs with Cr-silicide contacts, along with a field-effect hole mobility (μhFE) of 25 cm2/Vs. To the best of our knowledge, the hole and electron mobilities reported here are the highest achieved to date using direct PECVD. © 2005 IEEE.
Resumo:
Occurrence and growth rates of two species of intertidal fouling bryozoans namely Electra bengalensis (Stoliczka) and Electra crustulanta (Pallas) are presented in this paper. The former was a typically marine form, settling on panels only during the high saline conditions of the pre-monsoon period and were absent during the low salinity conditions of the monsoon period, while the latter appeared to be a typical brackish water form settling on panels during the low saline conditions existing during the monsoon and post-monsoon periods and were totally absent during the pre-monsoon months. Regression co-efficient of the former was higher than that of the latter suggesting more pronounced growth in Electra bengalensis. Maximum growth for this species was noticed during March, April and May (pre-monsoon) while for the other species growth was more or less similar during monsoon and post-monsoon months (June-January) showing that the species was at home in oligohaline and mesohaline waters.
Resumo:
FIRRI surveyed the fisheries of Lake George and Kazinga Channel between 20th June and 20th July 2001. This was the second survey FIRRI has conducted for the ILM project on the water system. The first survey was conducted during November 2000. These data, the analyses and accompanying reports contribute to baseline information for the fishery being collected with the support of ILM that is required for lakewide planning and management. Eight fish landing sites (6 on Lake George) namely; Kahendero, Hamukungu, Kasenyi, Kashaka, Mahyoro, Kayinja (2 on Kazinga Channel) namely; Katunguru -K and B fall within the focus of ILM and were surveyed during November 2000 and June/July 2001 over a three day period at each landing site in 2001 (Mahyoro 2 days). In November 2000, each landing was sampled once. FIRRI conducted a rapid FS and concurrently a CAS. All results are reported by landing site and then summed up (Global) for 8 sites on Lake George and Kazinga Channel.
Resumo:
Acid-sensing ion channels (ASICs) are emerging as fundamental players in the regulation of neural plasticity and in pathological conditions. Here we showed that lead (Pb2+), a well known neurotoxic metal ion, reversibly and concentration-dependently inhib
Resumo:
Brachyuran larvae are the most common zooplankton component of the Manora Channel, Karachi, Pakistan. The identification of these larvae would assist in the assessment of brachyuran species and provide information on percentage composition, occurrence, abundance and breeding habits of the brachyuran species in the Manora Channel area. However plankton caught larvae is not easily identified. An accurate identification of such material is only possible by the comparison with larvae reared under laboratory conditions and documented with illustrations. The identifications for this present study were based on the works of Gurney (1938); Atkins (1954); Chhapgar (1955); Raja Bai (1960); Hashmi (1969, 1970a, b); Baba and Miyata (1971); Kakati and Sankolli (1975); Rice (1975); Kakati (1977); Lim and Tan (1981); Yatsuzuka and Sakai (1984); Fielder, et al (1984); Amir (1989, M. Phil thesis unpublished); Ingle (1992); Siddiqui and Tirmizi (1992); Tirmizi et al (1993); Bano (1999, M. Phil thesis unpublished); Ghory and Siddiqui (2001); Ghory (2002, M. Phil thesis unpublished); Ghory and Siddiqui (2002).
Resumo:
This study gives an account of distributional patterns of Brachyuran larvae in the Manora Channel from January to November 1995. The planktonic sampling was carried out during day time from surface and sub-surface waters of station I and II (certain sites) at shallow depths (15'-20') using Bongo net of 300 micron mesh size. In all 19527 larvae were obtained through fourteen sampling. These brachyuran larvae belonged to nine families and twenty four species: Ebalia sagittifera, Philyra sp., Philyra scabriuscula (Leucosiidae), Schizophyris aspera (Majidae), Charybdis annulata, Charybdis sp. (Portunidae), Xanthid sp A., B. and C. (Xanthidae), Pilumnus karachiensis, Pilumnus sp. (Pilumnidae), Menippe rumphii (Oziidae), Pinnotheres sp. A, and B. (Pinnotheridae), Nasima dotilliforme, Serenella indica, Macrophthalmus (Mareotis) depressus, Macrophthalmus sp., Dotilla blanfordi, Ocypodid sp. A., B. and C. (Ocypodidae), Metopograpsus thukuhar and Clistocoeloma lanatum (Grapsidae). This study is based on identification, occurrence, distributional patterns along Manora Channel and percentage composition of brachyuran larvae in the area, collected during 1995.