Sciweavers

793 search results - page 89 / 159
» Confidence-weighted linear classification
Sort
View
DMIN
2006
158views Data Mining» more  DMIN 2006»
15 years 1 months ago
Ensemble Selection Using Diversity Networks
- An ideal ensemble is composed of base classifiers that perform well and that have minimal overlap in their errors. Eliminating classifiers from an ensemble based on a criterion t...
Qiang Ye, Paul W. Munro
NIPS
2001
15 years 1 months ago
Kernel Machines and Boolean Functions
We give results about the learnability and required complexity of logical formulae to solve classification problems. These results are obtained by linking propositional logic with...
Adam Kowalczyk, Alex J. Smola, Robert C. Williamso...
AMC
2008
81views more  AMC 2008»
15 years 9 hour ago
On Steepest-Descent-Kaczmarz methods for regularizing systems of nonlinear ill-posed equations
We investigate modified steepest descent methods coupled with a loping Kaczmarz strategy for obtaining stable solutions of nonlinear systems of ill-posed operator equations. We sh...
A. De Cezaro, Markus Haltmeier, Antonio Leit&atild...
AMC
2010
114views more  AMC 2010»
15 years 9 hour ago
Witness sets of projections
Elimination is a basic algebraic operation which geometrically corresponds to projections. This article describes using the numerical algebraic geometric concept of witness sets t...
Jonathan D. Hauenstein, Andrew J. Sommese
DM
2010
144views more  DM 2010»
14 years 12 months ago
Balanced Cayley graphs and balanced planar graphs
A balanced graph is a bipartite graph with no induced circuit of length 2 (mod 4). These graphs arise in linear programming. We focus on graph-algebraic properties of balanced gra...
Joy Morris, Pablo Spiga, Kerri Webb