Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures

Osama Waqar Bhatti; Nikita Ambasana; Madhavan Swaminathan

doi:10.1109/IMS19712.2021.9574988

Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures

Osama Waqar Bhatti, Nikita Ambasana, Madhavan Swaminathan

Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

Current deterministic machine learning models provide point estimates for predictions without any metric quantifying its inaccuracy for test inputs. In this paper, we focus on uncertainty analysis for a recently developed machine learning model used for design space and frequency response extrapolation using variational inference. This information equips the designer to identify how well the model performs for a given test input and hence identify if further training is required. We also explain here how much data is enough to train this model well. We discuss these approaches for a 5th order interdigital bandpass filter at 28GHz.

Original language	English (US)
Title of host publication	2021 IEEE MTT-S International Microwave Symposium, IMS 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	16-19
Number of pages	4
ISBN (Electronic)	9781665403078
DOIs	https://doi.org/10.1109/IMS19712.2021.9574988
State	Published - Jun 7 2021
Event	2021 IEEE MTT-S International Microwave Symposium, IMS 2021 - Virtual, Atlanta, United States Duration: Jun 7 2021 → Jun 25 2021

Publication series

Name	IEEE MTT-S International Microwave Symposium Digest
Volume	2021-June
ISSN (Print)	0149-645X

Conference

Conference	2021 IEEE MTT-S International Microwave Symposium, IMS 2021
Country/Territory	United States
City	Virtual, Atlanta
Period	6/7/21 → 6/25/21

All Science Journal Classification (ASJC) codes

Radiation
Condensed Matter Physics
Electrical and Electronic Engineering

Access to Document

10.1109/IMS19712.2021.9574988

Cite this

Bhatti, O. W., Ambasana, N., & Swaminathan, M. (2021). Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures. In 2021 IEEE MTT-S International Microwave Symposium, IMS 2021 (pp. 16-19). (IEEE MTT-S International Microwave Symposium Digest; Vol. 2021-June). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IMS19712.2021.9574988

Bhatti, Osama Waqar ; Ambasana, Nikita ; Swaminathan, Madhavan. / Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures. 2021 IEEE MTT-S International Microwave Symposium, IMS 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 16-19 (IEEE MTT-S International Microwave Symposium Digest).

@inproceedings{8e01c36fee794c0ca315be4a2b3f2312,

title = "Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures",

abstract = "Current deterministic machine learning models provide point estimates for predictions without any metric quantifying its inaccuracy for test inputs. In this paper, we focus on uncertainty analysis for a recently developed machine learning model used for design space and frequency response extrapolation using variational inference. This information equips the designer to identify how well the model performs for a given test input and hence identify if further training is required. We also explain here how much data is enough to train this model well. We discuss these approaches for a 5th order interdigital bandpass filter at 28GHz.",

author = "Bhatti, {Osama Waqar} and Nikita Ambasana and Madhavan Swaminathan",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE MTT-S International Microwave Symposium, IMS 2021 ; Conference date: 07-06-2021 Through 25-06-2021",

year = "2021",

month = jun,

day = "7",

doi = "10.1109/IMS19712.2021.9574988",

language = "English (US)",

series = "IEEE MTT-S International Microwave Symposium Digest",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "16--19",

booktitle = "2021 IEEE MTT-S International Microwave Symposium, IMS 2021",

address = "United States",

}

Bhatti, OW, Ambasana, N & Swaminathan, M 2021, Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures. in 2021 IEEE MTT-S International Microwave Symposium, IMS 2021. IEEE MTT-S International Microwave Symposium Digest, vol. 2021-June, Institute of Electrical and Electronics Engineers Inc., pp. 16-19, 2021 IEEE MTT-S International Microwave Symposium, IMS 2021, Virtual, Atlanta, United States, 6/7/21. https://doi.org/10.1109/IMS19712.2021.9574988

Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures. / Bhatti, Osama Waqar; Ambasana, Nikita; Swaminathan, Madhavan.
2021 IEEE MTT-S International Microwave Symposium, IMS 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 16-19 (IEEE MTT-S International Microwave Symposium Digest; Vol. 2021-June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures

AU - Bhatti, Osama Waqar

AU - Ambasana, Nikita

AU - Swaminathan, Madhavan

PY - 2021/6/7

Y1 - 2021/6/7

N2 - Current deterministic machine learning models provide point estimates for predictions without any metric quantifying its inaccuracy for test inputs. In this paper, we focus on uncertainty analysis for a recently developed machine learning model used for design space and frequency response extrapolation using variational inference. This information equips the designer to identify how well the model performs for a given test input and hence identify if further training is required. We also explain here how much data is enough to train this model well. We discuss these approaches for a 5th order interdigital bandpass filter at 28GHz.

AB - Current deterministic machine learning models provide point estimates for predictions without any metric quantifying its inaccuracy for test inputs. In this paper, we focus on uncertainty analysis for a recently developed machine learning model used for design space and frequency response extrapolation using variational inference. This information equips the designer to identify how well the model performs for a given test input and hence identify if further training is required. We also explain here how much data is enough to train this model well. We discuss these approaches for a 5th order interdigital bandpass filter at 28GHz.

UR - http://www.scopus.com/inward/record.url?scp=85118532491&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85118532491&partnerID=8YFLogxK

U2 - 10.1109/IMS19712.2021.9574988

DO - 10.1109/IMS19712.2021.9574988

M3 - Conference contribution

AN - SCOPUS:85118532491

T3 - IEEE MTT-S International Microwave Symposium Digest

SP - 16

EP - 19

BT - 2021 IEEE MTT-S International Microwave Symposium, IMS 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 IEEE MTT-S International Microwave Symposium, IMS 2021

Y2 - 7 June 2021 through 25 June 2021

ER -

Bhatti OW, Ambasana N, Swaminathan M. Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures. In 2021 IEEE MTT-S International Microwave Symposium, IMS 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 16-19. (IEEE MTT-S International Microwave Symposium Digest). doi: 10.1109/IMS19712.2021.9574988

Machine Learning Based Uncertainty Quantification of Extrapolated Design Space and Frequency Response for RF Structures

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this