(Enter summary)
Abstract: AN EMPIRICAL APPROACH TO REAL-VALUED
MULTIPLE-INSTANCE CLASSIFICATIONS
by Robert A. Amar
ADVISOR: Professor Sally A. Goldman
August, 2000
St. Louis, Missouri
The model of multiple-instance learning has recently arisen in the machine learning
community as a new and important setting for several real-world problems, most notably
drug design and discovery. However, most work, both theoretical and empirical, has
been aimed at classifying target objects in a Boolean, or binary, fashion. Often we
... (Update)
Similar documents based on text: More All
0.4: Multiple Instance Regression - Ray, Page (2001)
(Correct)
0.3: EM-DD: An Improved Multiple-Instance Learning Technique - Zhang, Goldman (2001)
(Correct)
0.3: Multiple Instance Learning with Generalized Support.. - Andrews, Hofmann.. (2002)
(Correct)
BibTeX entry: (Update)
@misc{ amar-empirical,
author = "Robert A. Amar",
title = "An Empirical Approach To Real-Valued Multiple-Instance Classifications",
url = "citeseer.ist.psu.edu/amar00empirical.html" }
Citations (may not include all citations):
79
Solving the multiple-instance problem with axis-parallel rec..
- Dietterich, Lathrop et al. - 1997
27
A framework for multiple-instance learning
- Maron, Lozano-Prez - 1998
22
On learning from multi-instance examples: Empirical evaluati..
- Auer - 1997
17
Learning from ambiguity
- Maron - 1998
Documents on the same site (http://siesta.cs.wustl.edu/cs/cs/publications.html): More
Recognition and Verification of Design Patterns - Plezbert, Cytron (2000)
(Correct)
CodeWeave: Exploring Fine-Grained Mobility of Code - Mascolo, Picco, Roman (2000)
(Correct)
Error Control For Continuous Media And Large Scale Multicast.. - Papadopoulos (1999)
(Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC