We consider the problem of learning perturbation-based probabilistic models by computing and differentiating expected losses. This is a challenging computational problem that has traditionally been tackled using Monte Carlo-based methods. In this work, we show how a generalization of parametric min-cuts can be used to address the same problem, achieving higher accuracy and faster performance th...

iranian female participation rate in the labor market of iran in the recent decade shows a increasing trend which amounted from 8% in 1999 to 18% in 2005.using parametric and non-parametric econometrics methods, in this research we aim to study the behavior of individual married females labor participation. our dataset consist of the annually budget survey of 2006 which is reported by statistic...

Since the inception of intuitionistic fuzzy sets in 1986, many authors have proposed different methods for ranking intuitionistic fuzzy numbers (IFNs). How ever, due to the complexity of the problem, a method which gives a satisfactory result to all situations is a challenging task. Most of them contained some shortcomings, such as requirement of complicated calculations, inconsistency with hum...

In this paper we propose a methodology to measure the characteristics and composition of intermediate products using productivity indicators based on directional distance functions. We evaluate how quality attributes interact with the quantity level in grapes production, and find evidence of a trade-off between quantity and aggregate quality for Chardonnay. ∗Dipartimento di Scienze Economiche —...

This paper presents a novel, generic, scalable, autonomous, and flexible supervised learning algorithm for the classification of multivariate and variable length time series. The essential ingredients of the algorithm are randomization, segmentation of time-series, decision tree ensemble based learning of subseries classifiers, combination of subseries classification by voting, and cross-valida...

In this paper, we investigate image segmentation by region merging. Given any similarity measure between regions, satisfying some weak constraints, we give a general predicate for answering if two regions are to be merged or not during the segmentation process. Our predicate is generic and has six properties. The first one is its independance with respect to the similarity measure, that leads t...

This report present a non parametric matching method based on non parametric invariants to geometry and intensity transforms of images. The use of invariant makes it possible to match images under large geometric transforms. The use of non parametric measures makes it robust to partial occlusion. An algorithm to obtain sub-pixel precision is also shown. Results on real images validate the appro...