Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

22

ECML
2004
Springer

favoriteEmaildiscussreport

163views Machine Learning» more ECML 2004»

A Boosting Approach to Multiple Instance Learning

13 years 10 months ago

A Boosting Approach to Multiple Instance Learning

Download sci2s.ugr.es

In this paper we present a boosting approach to multiple instance learning. As weak hypotheses we use balls (with respect to various metrics) centered at instances of positive bags. For the ∞-norm these hypotheses can be modiﬁed into hyper-rectangles by a greedy algorithm. Our approach includes a stopping criterion for the algorithm based on estimates for the generalization error. These estimates can also be used to choose a preferable metric and data normalization. Compared to other approaches our algorithm delivers improved or at least competitive results on several multiple instance benchmark data sets.

Peter Auer, Ronald Ortner

Real-time Traffic

ECML 2004 | Multiple Instance | Multiple Instance Learning | Weak Hypotheses |

claim paper

Related Content

» Online Semisupervised MultipleInstance Boosting

» MultipleInstance Learning via Disjunctive Programming Boosting

» Multiple instance tracking based on hierarchical maximizing bags margin boosting

» Multiple Instance Boost Using Graph Embedding Based Decision Stump for Pedestrian Detectio...

» Multiple Instance Boosting for Face Recognition in Videos

» Multiple Instance Feature for Robust Partbased Object Detection

» Inverse Multiple Instance Learning for Classifier Grids

» MIForests MultipleInstance Learning with Randomized Trees

» Revisiting MultipleInstance Learning Via Embedded Instance Selection

Post Info
More Details (n/a)

Added	01 Jul 2010
Updated	01 Jul 2010
Type	Conference
Year	2004
Where	ECML
Authors	Peter Auer, Ronald Ortner

Comments (0)