1,336 research outputs found
Multivariate L1 Statistical Methods: The Package MNM
No full textIn the paper we present an R package MNM dedicated to multivariate data analysis based on the L_1 norm. The analysis proceeds very much as does a traditional multivariate analysis. The regular L_2 norm is just replaced by different L_1 norms, observation vectors are replaced by their (standardized and centered) spatial signs, spatial ranks, and spatial signed-ranks, and so on. The procedures are fairly efficient and robust, and no moment assumptions are needed for asymptotic approximations. The background theory is briefly explained in the multivariate linear regression model case, and the use of the package is illustrated with several examples using the R package MNM.
Stability analysis of Oja-RLS learning rule
No full text. It is shown that the discrete time dynamical system defined by the Oja-RLS algorithm is stable in the closed ring K(0; 9=8) \Gamma K(0; 8=9) if only the initial gain fi 0 is bounded by (2B) \Gamma1 ; where B = b 2 and b is the bound for the learning sequence. It is rigorously proved that automatically computed gains fi n in Oja-RLS scheme converge to zero with the rate 1=n; almost surely. 1. Introduction Oja-RLS algorithm ([3]) is a neural type iterative scheme w n+1 = w n + fi n f(x n ; w n ) (1) used for stochastic approximation of a principal vector w 2 R N for the given sequence (x n ) of iid (independent identically distributed) random variables. The real positive number fi n is called the gain or the learning rate coefficient at time n: Contrary to the classical Oja rule, gains are not properly chosen constants, but they are random values which are computed recursively: 1 fi 0 = 2B; 1 fi n+1 = 1 fi n + y 2 n (2) The research of both authors was sponsored by E..
SAS Macros for Calculation of Population Attributable Fraction in a Cohort Study Design
No full textThe population attributable fraction (PAF) is a useful measure for quantifying the impact of exposure to certain risk factors on a particular outcome at the population level. Recently, new model-based methods for the estimation of PAF and its confidence interval for different types of outcomes in a cohort study design have been proposed. In this paper, we introduce SAS macros implementing these methods and illustrate their application with a data example on the impact of different risk factors on type 2 diabetes incidence.
Learning Compatibitlity Functions for Feature Binding and Perceptual Grouping
No full textWeng S, Steil JJ. Learning Compatibitlity Functions for Feature Binding and Perceptual Grouping. In: Kaynak O, Alpaydin E, Oja E, Xu L, eds. Proc. of Int. Conference Artificial Neural Networks. Vol LNCS 2714. Springer Verlag; 2003: 60-67
New perspectives on computer concepts, 2010, comprehensive / Parsons, Oja.
No full textAccompanying BookOnCD contains the entire contents of the textbook with figures that come to life as videos, software tours, and animations. Includes also labs and computer-scored practice tests.Includes indexxxiii, 749 pages. :Get Synched! Learning happens when students and instructors effectively connect and communicate. NEW PERSPECTIVES ON COMPUTER CONCEPTS 2011 makes it possible. Creating a fully integrated and interactive teaching and learning environment, NEW PERSPECTIVES ON COMPUTER CONCEPTS 2011 consists of a printed book, companion Web site, media rich BookOnCD, and WebTrack assessment. Students and Instructors work in sync to explore, understand, and apply computer technology in everyday life
Active Learning in Self-Organizing Maps
No full textHasenjäger M, Ritter H, Obermayer K. Active Learning in Self-Organizing Maps. In: Oja E, Kaski S, eds. Kohonen Maps. Amsterdam: Elsevier; 1999: 57-70
Unsupervised Learning of a Kinematic Arm Model
No full textHoffmann H, Möller R. Unsupervised Learning of a Kinematic Arm Model. In: Kaynak O, Alpaydin E, Oja E, Xu L, eds. Artificial Neural Networks and Neural Information Processing –- ICANN/ICONIP 2003. Lecture Notes in Computer Sciences 2714. Springer; 2003: 463-470
Barry-Shaw, Nikolas and Dru Oja Jay. 2012. Paved with Good Intentions: Canada\u27s Development NGOs from Idealism to Imperialism.
No full textBarry-Shaw, Nikolas and Dru Oja Jay. 2012. Paved with Good Intentions: Canada\u27s Development NGOs from Idealism to Imperialism
Estado ecológico del río Oja (cuenca del Ebro, La Rioja, España), mediante indicadores de macroinvertebrados
No full textEl presente trabajo presenta resultados sobre el estado ecológico del río Oja o Glera (cuenca del Oja- Tirón) durante los años 2003-2007, así como información proveniente de datos históricos recopilados a partir de los informes de cuenca existentes (informe de 2004, que incluye datos desde 1994 hasta 2004 e informe de 2007). Se han seleccionado 7 puntos, distribuidos entre las tres tipologías definidas en el río Oja: en el ecotipo 11, ríos de montaña mediterránea silícea, 3 puntos; en el ecotipo 26, ríos de montaña húmeda calcárea, 1 punto y en el ecotipo 12, Ríos de montaña mediterránea calcárea, 3 puntos. Para cada punto se han calculado una serie de índices de macroinvertebrados (1BMWP' 1ASPT, NFAM, NFPLE y EPT) Y sus ratios de estado ecológico (EQR). En conjunto, el 75.86 % de las medidas del1BMWP dan una calidad Muy buena o Buena y un 13.79 % Moderada mientras que el 10.35 % aparece como Deficiente. En cuanto a su evolución histórica, en aquellos tramos del río que presentan agua de manera permanente se observó una mejora progresiva de la calidad del agua desde los primeros datos de los que se dispone (años 1994-95) hasta los más recientes (2007), según los valores más recientes de clasificación de calidad de agua para el1BMWP (BOE, 2008).This work provides results about the water quality of the Oja or Glera River (Oja- Tirón basin) during the years 2003-2007, as well as information from the historical data taken from previous basin reports (2004 report, that includes data from 1994 to 2004 and 2007 report). Seven sampling points have been selected, distributed between the three typologies defined in the Oja River: in ecotype ll, Siliceous Mediterranean mountain river, 3 points; in ecotype 26, Humid calcareous mountain river, 1 point and in ecotype 12, Mediterranean calcareous mountain rivers, 3 points. Several macro invertebrate indexes (IBMWP, IASPT, NFAM, NFPLE and EPT) and their ecological quality ratios (EQR) were calculated for each point. As a whole, 75.86 % of the IBMWP measurements show Very good or Good quality, with a 13.79 % of Moderate quality, meanwhile 10.35 % of them appear as Deficient. Regarding its historie evolution, in those stretches of the river which present permanent water a progressive improvement of water quality was observed from the first available data (years 1994-95) until the most recent ones (2007), according to the most recent values of water quality classification for IBMWP (BOE, 2008)
- …
