Multiple instance learning with differential evolutionary pooling

Kamanasish Bhattacharjee; Arti Tiwari; Millie Pant; Chang Wook Ahn; Sanghoun Oh

doi:10.3390/electronics10121403

Multiple instance learning with differential evolutionary pooling

Kamanasish Bhattacharjee, Arti Tiwari, Millie Pant, Chang Wook Ahn, Sanghoun Oh

Research output: Contribution to journal › Article › peer-review

Abstract

While implementing Multiple Instance Learning (MIL) through Deep Neural Networks, the most important task is to design the bag-level pooling function that defines the instance-to-bag relationship and eventually determines the class label of a bag. In this article, Differential Evolutionary (DE) pooling—an MIL pooling function based on Differential Evolution (DE) and a bio-inspired metaheuristic—is proposed for the optimization of the instance weights in parallel with training the Deep Neural Network. This article also presents the effects of different parameter adaptation techniques with different variants of DE on MIL.

Original language	English
Article number	1403
Journal	Electronics (Switzerland)
Volume	10
Issue number	12
DOIs	https://doi.org/10.3390/electronics10121403
State	Published - 2 Jun 2021
Externally published	Yes

Keywords

Adaptive
Differential Evolution (DE)
Multiple Instance Learning (MIL)
Parameter
Pooling
Variant

ASJC Scopus subject areas

Control and Systems Engineering
Signal Processing
Hardware and Architecture
Computer Networks and Communications
Electrical and Electronic Engineering

Access to Document

10.3390/electronics10121403

Cite this

@article{bc2948b91e724ca899ed30d9abe9543c,

title = "Multiple instance learning with differential evolutionary pooling",

abstract = "While implementing Multiple Instance Learning (MIL) through Deep Neural Networks, the most important task is to design the bag-level pooling function that defines the instance-to-bag relationship and eventually determines the class label of a bag. In this article, Differential Evolutionary (DE) pooling—an MIL pooling function based on Differential Evolution (DE) and a bio-inspired metaheuristic—is proposed for the optimization of the instance weights in parallel with training the Deep Neural Network. This article also presents the effects of different parameter adaptation techniques with different variants of DE on MIL.",

keywords = "Adaptive, Differential Evolution (DE), Multiple Instance Learning (MIL), Parameter, Pooling, Variant",

author = "Kamanasish Bhattacharjee and Arti Tiwari and Millie Pant and Ahn, {Chang Wook} and Sanghoun Oh",

note = "Publisher Copyright: {\textcopyright} 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses /by/4.0/).",

year = "2021",

month = jun,

day = "2",

doi = "10.3390/electronics10121403",

language = "English",

volume = "10",

journal = "Electronics (Switzerland)",

issn = "2079-9292",

publisher = "MDPI Multidisciplinary Digital Publishing Institute",

number = "12",

}

TY - JOUR

T1 - Multiple instance learning with differential evolutionary pooling

AU - Bhattacharjee, Kamanasish

AU - Tiwari, Arti

AU - Pant, Millie

AU - Ahn, Chang Wook

AU - Oh, Sanghoun

N1 - Publisher Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses /by/4.0/).

PY - 2021/6/2

Y1 - 2021/6/2

N2 - While implementing Multiple Instance Learning (MIL) through Deep Neural Networks, the most important task is to design the bag-level pooling function that defines the instance-to-bag relationship and eventually determines the class label of a bag. In this article, Differential Evolutionary (DE) pooling—an MIL pooling function based on Differential Evolution (DE) and a bio-inspired metaheuristic—is proposed for the optimization of the instance weights in parallel with training the Deep Neural Network. This article also presents the effects of different parameter adaptation techniques with different variants of DE on MIL.

AB - While implementing Multiple Instance Learning (MIL) through Deep Neural Networks, the most important task is to design the bag-level pooling function that defines the instance-to-bag relationship and eventually determines the class label of a bag. In this article, Differential Evolutionary (DE) pooling—an MIL pooling function based on Differential Evolution (DE) and a bio-inspired metaheuristic—is proposed for the optimization of the instance weights in parallel with training the Deep Neural Network. This article also presents the effects of different parameter adaptation techniques with different variants of DE on MIL.

KW - Adaptive

KW - Differential Evolution (DE)

KW - Multiple Instance Learning (MIL)

KW - Parameter

KW - Pooling

KW - Variant

UR - http://www.scopus.com/inward/record.url?scp=85107469893&partnerID=8YFLogxK

U2 - 10.3390/electronics10121403

DO - 10.3390/electronics10121403

M3 - Article

AN - SCOPUS:85107469893

SN - 2079-9292

VL - 10

JO - Electronics (Switzerland)

JF - Electronics (Switzerland)

IS - 12

M1 - 1403

ER -

Multiple instance learning with differential evolutionary pooling

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this