What Can Machine Learning Do for Circuits?: Practical findings from principled machine learning applications
IEEE Solid-State Circuits Magazine · 2026 · 18 (2) · 83-92
DOI: 10.1109/mssc.2026.3680404 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Solid-State Circuits Magazine, 2026. Verfasst von Felicia B. Guo und 3 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben.
Abstract (Originalsprache)
Analog and mixed-signal circuit design automation has remained a longstanding objective spanning several decades of research. This article examines how machine learning (ML) can help realize new capabilities in this domain. AstRL [analog and mixed-signal (AMS) synthesis with reinforcement learning], a topology-level circuit design framework, is introduced to demonstrate recent progress in deep learning-based approaches toward automation. We then position AstRL within a broader survey of ML approaches for circuit design. From these examples, we distill several key development principles, including the importance of formulating ML methods that reflect realistic design challenges, leveraging simulation-guided training, and integrating expert knowledge. These core ideas further highlight a principled path toward applying modern ML architectures that are based on practical design considerations.
Zitieren
APA
Guo, F. B., Ho, K. T., Vladimirescu, A., & Nikolić, B. (2026). What Can Machine Learning Do for Circuits?: Practical findings from principled machine learning applications. IEEE Solid-State Circuits Magazine, 18(2), 83-92. https://doi.org/10.1109/mssc.2026.3680404
Vancouver
Guo FB, Ho KT, Vladimirescu A, Nikolić B. What Can Machine Learning Do for Circuits?: Practical findings from principled machine learning applications. IEEE Solid-State Circuits Magazine. 2026;18(2):83-92. doi:10.1109/mssc.2026.3680404
BibTeX
@article{Guo2026,
author = {Guo, Felicia B. and Ho, Ken T. and Vladimirescu, Andrei and Nikolić, Borivoje},
title = {What Can Machine Learning Do for Circuits?: Practical findings from principled machine learning applications},
journal = {IEEE Solid-State Circuits Magazine},
year = {2026},
volume = {18},
number = {2},
pages = {83--92},
doi = {10.1109/mssc.2026.3680404}
}
EUVIMED