In this paper we present a new analysis of two algorithms, Gradient Descent and Exponentiated Gradient, for solving regression problems in the on-line framework. Both these algorithms compute a prediction that depends linearly on the current instance, and then update the coeecients of this linear combination according to the gradient of the loss function. However, the two algorithms have distin...

Traditional text detection methods mostly focus on quadrangle text. In this study we propose a novel method named sliding line point regression (SLPR) in order to detect arbitrary-shape text in natural scene. SLPR regresses multiple points on the edge of text line and then utilizes these points to sketch the outlines of the text. The proposed SLPR can be adapted to many object detection archite...

This paper is about the experience gained in IBM Haifa Research Lab (HRL) in creating regression-suites and minimizing their size, while maintaining high quality as measured by coverage. The problem that we solve, while similar to the one addressed in the literature, has a key difference; the compaction algorithm has to be implemented on-line due to the large number of tests processed. We compa...

We evaluated and compared various broken-line regression models and SAS (SAS Inst. Inc., Cary, NC) procedures for estimating nutrient requirements from nutrient dose response data. We used the SAS (Version 9) procedures NLIN and NLMixed and the response data of Parr et al. (2003), who evaluated the isoleucine requirement of growing swine. The SAS NLIN was used to fit 2 different broken-line reg...

BACKGROUND The aim of this study was to assess the sigmoid line of elasticity in the human aorta. METHODS AND RESULTS The pressure-diameter relation was measured in the descending aorta in 120 subjects. In an additional group of 6 subjects, transient vena caval occlusion produced 5 sets of pressure-diameter data. We found that the best fit curve of the pooled pressure-diameter data was a thir...

One crucial activity in software product line (SPL) testing is the detection of erroneous artifact interactions when combined for a variant. This detection is similar to the retest test-case selection problem in regression testing, where change impact analysis is applied to reason about changed dependencies to be retested. In this paper, we propose automated change impact analysis based on incr...

Nowadays, automatic signature verification is an active area of researches in numerous applications such as bank check verification, access restriction or special areas such as police investigations. In our researches signature was captured by Topaz SigLite T-LBK750-HSB device, where some dynamic features of signature can be also registered. In many transactions, the electronic verification of ...

We consider the problem of on-line prediction of real-valued labels, assumed bounded in absolute value by a known constant, of new objects from known labeled objects. The prediction algorithm’s performance is measured by the squared deviation of the predictions from the actual labels. No stochastic assumptions are made about the way the labels and objects are generated. Instead, we are given a ...

Construction of simultaneous confidence bands for a percentile line in linear regression has been considered by several authors. But only conservative symmetric bands, which use critical constants over the whole covariate range (−∞, ∞), are available in the literature. New methods allow the construction of exact symmetric bands for a percentile line over a finite interval of the covariate x. Th...

Segmented line regression has been used in many applications, and the problem of estimating the number of change-points in segmented line regression has been discussed in Kim et al. (2000). This paper studies asymptotic properties of the number of change-points selected by the permutation procedure of Kim et al. (2000). This procedure is based on a sequential application of likelihood ratio typ...