Search results for "Bayes"
showing 10 items of 847 documents
Towards Open Domain Chatbots—A GRU Architecture for Data Driven Conversations
2018
Understanding of textual content, such as topic and intent recognition, is a critical part of chatbots, allowing the chatbot to provide relevant responses. Although successful in several narrow domains, the potential diversity of content in broader and more open domains renders traditional pattern recognition techniques inaccurate. In this paper, we propose a novel deep learning architecture for content recognition that consists of multiple levels of gated recurrent units (GRUs). The architecture is designed to capture complex sentence structure at multiple levels of abstraction, seeking content recognition for very wide domains, through a distributed scalable representation of content. To …
IceCube search for neutrinos coincident with compact binary mergers from LIGO-Virgo's first gravitational-wave transient catalog
2020
Using the IceCube Neutrino Observatory, we search for high-energy neutrino emission coincident with compact binary mergers observed by the LIGO and Virgo gravitational-wave (GW) detectors during their first and second observing runs. We present results from two searches targeting emission coincident with the sky localization of each GW event within a 1000 s time window centered around the reported merger time. One search uses a model-independent unbinned maximum-likelihood analysis, which uses neutrino data from IceCube to search for pointlike neutrino sources consistent with the sky localization of GW events. The other uses the Low-Latency Algorithm for Multi-messenger Astrophysics, which …
Multioutput Automatic Emulator for Radiative Transfer Models
2018
This paper introduces a methodology to construct emulators of costly radiative transfer models (RTMs). The proposed methodology is sequential and adaptive, and it is based on the notion of acquisition functions in Bayesian optimization. Here, instead of optimizing the unknown underlying RTM function, one aims to achieve accurate approximations. The Automatic Multi-Output Gaussian Process Emulator (AMO-GAPE) methodology combines the interpolation capabilities of Gaussian processes (GPs) with the accurate design of an acquisition function that favors sampling in low density regions and flatness of the interpolation function. We illustrate the promising capabilities of the method for the const…
Efficient remote sensing image classification with Gaussian processes and Fourier features
2017
This paper presents an efficient methodology for approximating kernel functions in Gaussian process classification (GPC). Two models are introduced. We first include the standard random Fourier features (RFF) approximation into GPC, which largely improves the computational efficiency and permits large scale remote sensing data classification. In addition, we develop a novel approach which avoids randomly sampling a number of Fourier frequencies, and alternatively learns the optimal ones using a variational Bayes approach. The performance of the proposed methods is illustrated in complex problems of cloud detection from multispectral imagery.
GIGJ: a crustal gravity model of the Guangdong Province for predicting the geoneutrino signal at the JUNO experiment
2019
Gravimetric methods are expected to play a decisive role in geophysical modeling of the regional crustal structure applied to geoneutrino studies. GIGJ (GOCE Inversion for Geoneutrinos at JUNO) is a 3D numerical model constituted by ~46 x 10$^{3}$ voxels of 50 x 50 x 0.1 km, built by inverting gravimetric data over the 6{\deg} x 4{\deg} area centered at the Jiangmen Underground Neutrino Observatory (JUNO) experiment, currently under construction in the Guangdong Province (China). The a-priori modeling is based on the adoption of deep seismic sounding profiles, receiver functions, teleseismic P-wave velocity models and Moho depth maps, according to their own accuracy and spatial resolution. …
Revisiting the Epipalaeolithic-Neolithic Transition in the Extreme NW of Africa : The Latest Results of the Chronological Sequence of the Cave of Kaf…
2021
[EN] This study focuses on the chronostratigraphic sequence of the Cave of Kaf Taht el-Ghar (Dar Ben Karrich, Tétouan, Morocco) excavated in 2012 in the framework of the AGRIWESTMED research project. The broad sequence reveals a series of occupations ranging from the Pleistocene (Moroccan Aterian) to recent historical times. Our research identifies a rich Early Neolithic phase (sixth millennium cal BC) containing the earliest pottery and domesticated animal and plant remains in the western Maghreb. However, this Early Neolithic level is not an immediate successor of the last traces of the Epipalaeolithic hunter-gatherer occupation, which started at the end of the Younger Dryas (10,900–9700 …
Non‐linearity in interspecific interactions in response to climate change: cod and haddock as an example
2020
Climate change has profound ecological effects, yet our understanding of how trophic interactions among species are affected by climate change is still patchy. The sympatric Atlantic haddock and cod are co-occurring across the North Atlantic. They compete for food at younger stages and thereafter the former is preyed by the latter. Climate change might affect the interaction and coexistence of these two species. Particularly, the increase in sea temperature (ST) has been shown to affect distribution, population growth and trophic interactions in marine systems. We used 33-year long time series of haddock and cod abundances estimates from two data sources (acoustic and trawl survey) to analy…
Using latent variable models to identify large networks of species‐to‐species associations at different spatial scales
2015
Summary We present a hierarchical latent variable model that partitions variation in species occurrences and co-occurrences simultaneously at multiple spatial scales. We illustrate how the parameterized model can be used to predict the occurrences of a species by using as predictors not only the environmental covariates, but also the occurrences of all other species, at all spatial scales. We leverage recent progress in Bayesian latent variable models to implement a computationally effective algorithm that enables one to consider large communities and extensive sampling schemes. We exemplify the framework with a community of 98 fungal species sampled in c. 22 500 dead wood units in 230 plot…
The legacy of a vanished sea: a high level of diversification within a European freshwater amphipod species complex driven by 15 My of Paratethys reg…
2016
16 pages; International audience; The formation of continental Europe in the Neogene was due to the regression of the Tethys Ocean and of the Paratethys Sea. The dynamic geology of the area and repetitious transitions between marine and freshwater conditions presented opportunities for the colonization of newly emerging hydrological networks and diversification of aquatic biota. Implementing mitochondrial and nuclear markers in conjunction with a large-scale sampling strategy, we investigated the impact of this spatiotemporal framework on the evolutionary history of a freshwater crustacean morphospecies. The Gammarus balcanicus species complex is widely distributed in the area previously oc…
Habitat Discontinuities Separate Genetically Divergent Populations of a Rocky Shore Marine Fish.
2016
Habitat fragmentation has been suggested to be responsible for major genetic differentiations in a range of marine organisms. In this study, we combined genetic data and environmental information to unravel the relative role of geography and habitat heterogeneity on patterns of genetic population structure of corkwing wrasse (Symphodus melops), a rocky shore species at the northern limit of its distribution range in Scandinavia. Our results revealed a major genetic break separating populations inhabiting the western and southern coasts of Norway. This genetic break coincides with the longest stretch of sand in the whole study area, suggesting habitat fragmentation as a major driver of genet…