Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model

Freddy Gabbay; Ido Parchomovsky; Itay Yonatanov; Mohammad Omri

doi:10.1109/iolts69666.2026.11633803

Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model

Freddy Gabbay^*
, Ido Parchomovsky
, Itay Yonatanov
, Mohammad Omri

^*Corresponding author for this work

The Institute of Applied Physics

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

Abstract

Accurate dynamic voltage drop (DVD) analysis is increasingly critical in advanced nodes, where higher densities, lower voltages, and complex packaging exacerbate power delivery challenges. Traditional simulations are computationally expensive and typically performed late in the design cycle, risking costly redesigns. We propose a lightweight ML-based DVD prediction model using multi-scale CNNs, fusion layers, and skip connections to capture spatial and hierarchical power grid features. The model uniquely incorporates package and grid inductance and supports both vectorless and vector-based inputs. Evaluated on a 16 nm RISC-V core, it achieves 88.9 -94.2% accuracy with 7 mV tolerance and over 25,000× faster runtime than commercial tools.

Original language	English
Title of host publication	Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331546854
DOIs	https://doi.org/10.1109/iolts69666.2026.11633803
State	Published - 2026
Event	32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026 - Polignano a Mare, Italy Duration: 1 Jul 2026 → 3 Jul 2026

Publication series

Name	Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026

Conference

Conference	32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026
Country/Territory	Italy
City	Polignano a Mare
Period	1/07/26 → 3/07/26

Bibliographical note

Publisher Copyright:
© 2026 IEEE.

Keywords

Dynamic voltage drop
EDA
Machine learning
Reliability

Access to Document

10.1109/iolts69666.2026.11633803

Cite this

Gabbay, F., Parchomovsky, I., Yonatanov, I., & Omri, M. (2026). Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model. In Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026 (Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/iolts69666.2026.11633803

Gabbay, Freddy ; Parchomovsky, Ido ; Yonatanov, Itay et al. / Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model. Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026. Institute of Electrical and Electronics Engineers Inc., 2026. (Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026).

@inproceedings{604b36c426004824b2d14224c599b6d1,

title = "Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model",

abstract = "Accurate dynamic voltage drop (DVD) analysis is increasingly critical in advanced nodes, where higher densities, lower voltages, and complex packaging exacerbate power delivery challenges. Traditional simulations are computationally expensive and typically performed late in the design cycle, risking costly redesigns. We propose a lightweight ML-based DVD prediction model using multi-scale CNNs, fusion layers, and skip connections to capture spatial and hierarchical power grid features. The model uniquely incorporates package and grid inductance and supports both vectorless and vector-based inputs. Evaluated on a 16 nm RISC-V core, it achieves 88.9 -94.2\% accuracy with 7 mV tolerance and over 25,000{\texttimes} faster runtime than commercial tools.",

keywords = "Dynamic voltage drop, EDA, Machine learning, Reliability",

author = "Freddy Gabbay and Ido Parchomovsky and Itay Yonatanov and Mohammad Omri",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026 ; Conference date: 01-07-2026 Through 03-07-2026",

year = "2026",

doi = "10.1109/iolts69666.2026.11633803",

language = "אנגלית",

series = "Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026",

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

booktitle = "Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026",

address = "ארצות הברית",

}

Gabbay, F, Parchomovsky, I, Yonatanov, I & Omri, M 2026, Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model. in Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026. Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026, Institute of Electrical and Electronics Engineers Inc., 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026, Polignano a Mare, Italy, 1/07/26. https://doi.org/10.1109/iolts69666.2026.11633803

Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model. / Gabbay, Freddy; Parchomovsky, Ido; Yonatanov, Itay et al.
Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026. Institute of Electrical and Electronics Engineers Inc., 2026. (Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026).

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

TY - GEN

T1 - Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model

AU - Gabbay, Freddy

AU - Parchomovsky, Ido

AU - Yonatanov, Itay

AU - Omri, Mohammad

PY - 2026

Y1 - 2026

N2 - Accurate dynamic voltage drop (DVD) analysis is increasingly critical in advanced nodes, where higher densities, lower voltages, and complex packaging exacerbate power delivery challenges. Traditional simulations are computationally expensive and typically performed late in the design cycle, risking costly redesigns. We propose a lightweight ML-based DVD prediction model using multi-scale CNNs, fusion layers, and skip connections to capture spatial and hierarchical power grid features. The model uniquely incorporates package and grid inductance and supports both vectorless and vector-based inputs. Evaluated on a 16 nm RISC-V core, it achieves 88.9 -94.2% accuracy with 7 mV tolerance and over 25,000× faster runtime than commercial tools.

AB - Accurate dynamic voltage drop (DVD) analysis is increasingly critical in advanced nodes, where higher densities, lower voltages, and complex packaging exacerbate power delivery challenges. Traditional simulations are computationally expensive and typically performed late in the design cycle, risking costly redesigns. We propose a lightweight ML-based DVD prediction model using multi-scale CNNs, fusion layers, and skip connections to capture spatial and hierarchical power grid features. The model uniquely incorporates package and grid inductance and supports both vectorless and vector-based inputs. Evaluated on a 16 nm RISC-V core, it achieves 88.9 -94.2% accuracy with 7 mV tolerance and over 25,000× faster runtime than commercial tools.

KW - Dynamic voltage drop

KW - EDA

KW - Machine learning

KW - Reliability

UR - https://www.scopus.com/pages/publications/105047685252

U2 - 10.1109/iolts69666.2026.11633803

DO - 10.1109/iolts69666.2026.11633803

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:105047685252

T3 - Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026

BT - Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026

Y2 - 1 July 2026 through 3 July 2026

ER -

Gabbay F, Parchomovsky I, Yonatanov I, Omri M. Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model. In Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026. Institute of Electrical and Electronics Engineers Inc. 2026. (Proceedings - 2026 IEEE 32nd International Symposium on On-Line Testing and Robust System Design, IOLTS 2026). doi: 10.1109/iolts69666.2026.11633803

Accelerated Dynamic Voltage Drop Prediction Using a Lightweight Machine Learning Model

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this