Search results for "ESTIMATOR"
showing 10 items of 313 documents
Channel Capacity in Psychovisual Deep-Nets: Gaussianization Versus Kozachenko-Leonenko
2020
In this work, we quantify how neural networks designed from biology using no statistical training have a remarkable performance in information theoretic terms. Specifically, we address the question of the amount of information that can be extracted about the images from the different layers of psychophysically tuned deep networks. We show that analytical approaches are not possible, and we propose the use of two empirical estimators of capacity: the classical Kozachenko-Lonenko estimator and a recent estimator based on Gaussianization. Results show that networks purely based on visual psychophysics are extremely efficient in two aspects: (1) the internal representation of these networks dup…
Noise estimation from digital step-model signal
2013
International audience; This paper addresses the noise estimation in the digital domain and proposes a noise estimator based on the step signal model. It is efficient for any distribution of noise because it does not rely only on the smallest amplitudes in the signal or image. The proposed approach uses polarized/directional derivatives and a nonlinear combination of these derivatives to estimate the noise distribution (e.g., Gaussian, Poisson, speckle, etc.). The moments of this measured distribution can be computed and are also calculated theoretically on the basis of noise distribution models. The 1D performances are detailed, and as our work is mostly dedicated to image processing, a 2D…
Extrapolation des inventaires de biodiversité incomplets : comment estimer au mieux le nombre d’espèces manquantes et prévoir l’effort additionnel d’…
2016
While approaching or even reaching total completeness remains the idealgoal of any species inventory, incomplete samplings are deemed to become more often the rule, that are significantly more difficult to cope with (such as (micro-) invertebrates, for example).However, once post-operated appropriately, incomplete samplings may provide farmore data than it would seem and it is indeed possible to obtain valuable extra information froma reliable mode of extrapolation of the so-called Species Accumulation Curve.In fact, numerous tools designed for the exploitation of incomplete samplings havealready been proposed in the literature but the concern, with these numerous different tools, isprecise…
Statistical consequences of the Devroye inequality for processes. Applications to a class of non-uniformly hyperbolic dynamical systems
2005
In this paper, we apply Devroye inequality to study various statistical estimators and fluctuations of observables for processes. Most of these observables are suggested by dynamical systems. These applications concern the co-variance function, the integrated periodogram, the correlation dimension, the kernel density estimator, the speed of convergence of empirical measure, the shadowing property and the almost-sure central limit theorem. We proved in \cite{CCS} that Devroye inequality holds for a class of non-uniformly hyperbolic dynamical systems introduced in \cite{young}. In the second appendix we prove that, if the decay of correlations holds with a common rate for all pairs of functio…
Teacher shortages, teacher contracts and their effect on education in Africa
2010
1001; To enhance primary enrolment rates, many African countries have launched large teacher recruitment programmes in recent years. Given tight budgetary constraints, teachers are no longer employed in civil service positions, but on the basis of (fixed term) contracts typically implying considerably lower salaries and a sharply reduced amount of professional training. We analyse the effect of this change on educational quality in Niger, Togo and Mali, on the basis of very informative data, which are comparable across these countries. We use a variety of estimation techniques, including a non-parametric estimation of quantile treatment effects. Our results demonstrate that contract teacher…
Extrapolation of the Species Accumulation Curve for Incomplete Species Samplings: A New Nonparametric Approach to Estimate the Degree of Sample Compl…
2015
9 pages; International audience; Incomplete species samplings are deemed to remain the common practice in those groups of animals, such as small or micro- invertebrates, with numerous species that often are more or less difficult to detect in the field. Thus, extrapolating the Species Accumulation Curve as far as possible beyond the actual sample size may thus serve as a useful (although imperfect) surrogate to the desired, but practically inaccessible, complete samplings. In this context, several kinds of theoretical or empirical models for the Species Accumulation Curve and also a lot of estimators of the asymptotic limit of the Curve (i.e. total species richness) have been proposed. The …
A Bayesian Reconstruction of a Historical Population in Finland, 1647–1850
2020
This article provides a novel method for estimating historical population development. We review the previous literature on historical population time-series estimates and propose a general outline to address the well-known methodological problems. We use a Bayesian hierarchical time-series model that allows us to integrate the parish-level data set and prior population information in a coherent manner. The procedure provides us with model-based posterior intervals for the final population estimates. We demonstrate its applicability by estimating the long-term development of Finlands population from 1647 onward and simultaneously place the country among the very few to have an annual popula…
Network Capacity Estimators Predicting QoE in HTTP Adaptive Streaming
2022
The aim of adaptive HTTP streaming technology is preserving the best possible video streaming quality for viewers in heterogeneous network conditions. This can be achieved by making multiple quality versions of the video available. Switching between versions during playback should be imperceptible and fluent. The decision about quality-level switching is typically based on network capacity estimation and buffer occupancy, which predict the risk of stalling. Since quality-level switching and stalling are directly evident to the user, they are often classified as influence factors of quality of experience (QoE). In this paper, we observe different network capacity estimators and buffer behavi…
Simplified model of mean double step (MDS) in human body movement
2012
In this paper we present a simplified and useful model of the human body movement based on the full gait cycle description, called the Mean Double Step (MDS). It enables the parameterization and simplification of the human movement. Furthermore it allows a description of the gait cycle by providing standardized estimators to transform the gait cycle into a periodical movement process. Moreover the method of simplifying the MDS model and its compression are demonstrated. The simplification is achieved by reducing the number of bars of the spectrum and I or by reducing the number of samples describing the MDS both in terms of reducing their computational burden and their resources for the dat…
Diffuse ultraviolet erythemal irradiance on inclined planes: a comparison of experimental and modeled data.
2009
Values of measured and modeled diffuse UV erythemal irradiance (UVER) for all sky conditions are compared on planes inclined at 40 degrees and oriented north, south, east and west. The models used for simulating diffuse UVER are of the geometric-type, mainly the Isotropic, Klucher, Hay, Muneer, Reindl and Schauberger models. To analyze the precision of the models, some statistical estimators were used such as root mean square deviation, mean absolute deviation and mean bias deviation. It was seen that all the analyzed models reproduce adequately the diffuse UVER on the south-facing plane, with greater discrepancies for the other inclined planes. When the models are applied to cloud-free con…