Multiple-instance learning with pairwise instance similarity

Yuan, Liming; Liu, Jiafeng; Tang, Xianglong

Download

Title: Multiple-instance learning with pairwise instance similarity

Subtitle:

Group publication title:

AMCS, Volume 24 (2014)

Abstract:

Multiple-Instance Learning (MIL) has attracted much attention of the machine learning community in recent years and many real-world applications have been successfully formulated as MIL problems. Over the past few years, several Instance Selection-based MIL (ISMIL) algorithms have been presented by using the concept of the embedding space. ; Although they delivered very promising performance, they often require long computation times for instance selection, leading to a low efficiency of the whole learning process. In this paper, we propose a simple and efficient ISMIL algorithm based on the similarity of pairwise instances within a bag. The basic idea is selecting from every training bag a pair of the most similar instances as instance prototypes and then mapping training bags into the embedding space that is constructed from all the instance prototypes. ; Thus, the MIL problem can be solved with the standard supervised learning techniques, such as support vector machines. Experiments show that the proposed algorithm is more efficient than its competitors and highly comparable with them in terms of classification accuracy. Moreover, the testing of noise sensitivity demonstrates that our MIL algorithm is very robust to labeling noise.

Publisher:

Zielona Góra: Uniwersytet Zielonogórski

Resource Identifier:

oai:zbc.uz.zgora.pl:78971

DOI:

10.2478/amcs-2014-0041

Pages:

567-577

Source:

AMCS, volume 24, number 3 (2014) ; click here to follow the link

Language:

eng

Rights:

Biblioteka Uniwersytetu Zielonogórskiego

Subject and keywords:

multiple-instance learning instance selection similarity support vector machines

Object collections:

Digital Library of Zielona Góra > Repository > Faculties > Faculty of Computer, Electrical and Control Engineering

Digital Library of Zielona Góra > Repository > Types of work > Articles

Digital Library of Zielona Góra > Repository > Scientific journals and UZ publishing series > International Journal of Applied Mathematics and Computer Science (AMCS)

Digital Library of Zielona Góra > Repository > Scientific journals and UZ publishing series > International Journal of Applied Mathematics and Computer Science (AMCS) > International Journal of Applied Mathematics and Computer Science (AMCS) (2014)

Last modified:

Apr 26, 2024

In our library since:

Apr 26, 2024

Number of object content hits:

All available object's versions:

https://www.zbc.uz.zgora.pl/publication/88782

Show description in RDF format:

RDF

Show description in OAI-PMH format:

OAI-PMH

International Journal of Applied Mathematics and Computer Science (AMCS)

Edition name	Date
Multiple-instance learning with pairwise instance similarity	Apr 26, 2024

Object

Title: Multiple-instance learning with pairwise instance similarity

Creator:

Date:

Resource Type:

Contributor:

Subtitle:

Group publication title:

Abstract:

Publisher:

Resource Identifier:

DOI:

Pages:

Source:

Language:

Rights:

Subject and keywords:

Object collections:

Last modified:

In our library since:

Number of object content hits:

All available object's versions:

Show description in RDF format:

Show description in OAI-PMH format:

Objects

Similar

similar in FBC

Contact details

Address

Phone

E-Mail