Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search

Cheng Gong; Yang Nan; Lie Meng Pang; Hisao Ishibuchi; Qingfu Zhang

doi:10.1109/SSCI52147.2023.10372012

Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search

Cheng Gong
, Yang Nan
, Lie Meng Pang
, Hisao Ishibuchi^*
, Qingfu Zhang^*

^*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

2 Citations (Scopus)

Abstract

Remarkable successes in deep learning have spurred significant growth in the field of neural architecture search (NAS), which is rapidly advancing as a promising technique for automating the design of network architecture. From an optimization standpoint, a NAS task for a given search space can be viewed as a multi-objective optimization problem (MOP) when considering multiple design criteria simultaneously (e.g., prediction accuracy, architecture complexity, hardware efficiency). However, whether a NAS problem is a multimodal multi-objective optimization problem or not (i.e., whether a single non-dominated solution in the objective space has multiple different neural network architectures or not) has not been examined in the literature. This presents an intriguing research question that merits further investigation. To fill this gap, we examine the multimodal nature of seven multi-objective NAS problems. By doing so, this work aims to help MOP researchers to better understand the characteristics of the multi-objective NAS problems. © 2023 IEEE.

Original language	English
Title of host publication	2023 IEEE Symposium Series on Computational Intelligence (SSCI)
Publisher	IEEE
Pages	1821-1828
ISBN (Electronic)	978-1-6654-3065-4
DOIs	https://doi.org/10.1109/SSCI52147.2023.10372012
Publication status	Published - Dec 2023
Event	2023 IEEE Symposium Series on Computational Intelligence, SSCI 2023 - Mexico City, Mexico Duration: 5 Dec 2023 → 8 Dec 2023

Publication series

Name	IEEE Symposium Series on Computational Intelligence, SSCI

Conference

Conference	2023 IEEE Symposium Series on Computational Intelligence, SSCI 2023
Place	Mexico
City	Mexico City
Period	5/12/23 → 8/12/23

Funding

This work was supported by National Natural Science Foundation of China (Grant No. 62250710163, 62250710682), Guangdong Provincial Key Laboratory (Grant No. 2020B121201001), the Program for Guangdong Introducing Innovative and Enterpreneurial Teams (Grant No. 2017ZT07X386), the Stable Support Plan Program of Shenzhen Natural Science Fund (Grant No. 20200925174447003), Shenzhen Science and Technology Program (Grant No. KQTD2016112514355531), the Research Grants Council of the Hong Kong Special Administrative Region, China (GRF Project No. CityU11215622), and Natural Science Foundation of China (Project No: 62276223).

Research Keywords

Multi-objective optimization
Multimodal multi-objective optimization
Neural architecture search

RGC Funding Information

RGC-funded

Access Output

10.1109/SSCI52147.2023.10372012

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{076617a039974e59b86315d59cb28391,

title = "Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search",

abstract = "Remarkable successes in deep learning have spurred significant growth in the field of neural architecture search (NAS), which is rapidly advancing as a promising technique for automating the design of network architecture. From an optimization standpoint, a NAS task for a given search space can be viewed as a multi-objective optimization problem (MOP) when considering multiple design criteria simultaneously (e.g., prediction accuracy, architecture complexity, hardware efficiency). However, whether a NAS problem is a multimodal multi-objective optimization problem or not (i.e., whether a single non-dominated solution in the objective space has multiple different neural network architectures or not) has not been examined in the literature. This presents an intriguing research question that merits further investigation. To fill this gap, we examine the multimodal nature of seven multi-objective NAS problems. By doing so, this work aims to help MOP researchers to better understand the characteristics of the multi-objective NAS problems. {\textcopyright} 2023 IEEE.",

keywords = "Multi-objective optimization, Multimodal multi-objective optimization, Neural architecture search",

author = "Cheng Gong and Yang Nan and Pang, \{Lie Meng\} and Hisao Ishibuchi and Qingfu Zhang",

year = "2023",

month = dec,

doi = "10.1109/SSCI52147.2023.10372012",

language = "English",

series = "IEEE Symposium Series on Computational Intelligence, SSCI ",

publisher = "IEEE",

pages = "1821--1828",

booktitle = "2023 IEEE Symposium Series on Computational Intelligence (SSCI)",

address = "United States",

note = "2023 IEEE Symposium Series on Computational Intelligence, SSCI 2023 ; Conference date: 05-12-2023 Through 08-12-2023",

}

Gong, C, Nan, Y, Pang, LM, Ishibuchi, H & Zhang, Q 2023, Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search. in 2023 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE Symposium Series on Computational Intelligence, SSCI , IEEE, pp. 1821-1828, 2023 IEEE Symposium Series on Computational Intelligence, SSCI 2023, Mexico City, Mexico, 5/12/23. https://doi.org/10.1109/SSCI52147.2023.10372012

Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search. / Gong, Cheng; Nan, Yang; Pang, Lie Meng et al.
2023 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE, 2023. p. 1821-1828 (IEEE Symposium Series on Computational Intelligence, SSCI ).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search

AU - Gong, Cheng

AU - Nan, Yang

AU - Pang, Lie Meng

AU - Ishibuchi, Hisao

AU - Zhang, Qingfu

PY - 2023/12

Y1 - 2023/12

N2 - Remarkable successes in deep learning have spurred significant growth in the field of neural architecture search (NAS), which is rapidly advancing as a promising technique for automating the design of network architecture. From an optimization standpoint, a NAS task for a given search space can be viewed as a multi-objective optimization problem (MOP) when considering multiple design criteria simultaneously (e.g., prediction accuracy, architecture complexity, hardware efficiency). However, whether a NAS problem is a multimodal multi-objective optimization problem or not (i.e., whether a single non-dominated solution in the objective space has multiple different neural network architectures or not) has not been examined in the literature. This presents an intriguing research question that merits further investigation. To fill this gap, we examine the multimodal nature of seven multi-objective NAS problems. By doing so, this work aims to help MOP researchers to better understand the characteristics of the multi-objective NAS problems. © 2023 IEEE.

AB - Remarkable successes in deep learning have spurred significant growth in the field of neural architecture search (NAS), which is rapidly advancing as a promising technique for automating the design of network architecture. From an optimization standpoint, a NAS task for a given search space can be viewed as a multi-objective optimization problem (MOP) when considering multiple design criteria simultaneously (e.g., prediction accuracy, architecture complexity, hardware efficiency). However, whether a NAS problem is a multimodal multi-objective optimization problem or not (i.e., whether a single non-dominated solution in the objective space has multiple different neural network architectures or not) has not been examined in the literature. This presents an intriguing research question that merits further investigation. To fill this gap, we examine the multimodal nature of seven multi-objective NAS problems. By doing so, this work aims to help MOP researchers to better understand the characteristics of the multi-objective NAS problems. © 2023 IEEE.

KW - Multi-objective optimization

KW - Multimodal multi-objective optimization

KW - Neural architecture search

UR - https://www.scopus.com/pages/publications/85182952027

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85182952027&origin=recordpage

U2 - 10.1109/SSCI52147.2023.10372012

DO - 10.1109/SSCI52147.2023.10372012

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - IEEE Symposium Series on Computational Intelligence, SSCI

SP - 1821

EP - 1828

BT - 2023 IEEE Symposium Series on Computational Intelligence (SSCI)

PB - IEEE

T2 - 2023 IEEE Symposium Series on Computational Intelligence, SSCI 2023

Y2 - 5 December 2023 through 8 December 2023

ER -

Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Few for Many: A Non-Pareto Approach for Many Objective Optimization

Cite this

Examination of the Multimodal Nature of Multi-Objective Neural Architecture Search

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Few for Many: A Non-Pareto Approach for Many Objective Optimization

Cite this