PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 18, 2026Electronics0 citationsOpen Access

Physics-Inspired Deep Learning and Bayesian Optimization for Surrogate Modeling of Nanosheet and Forksheet Transistors

View Full Paper
BSBakhita SalmanCMCamilla MancillasMYMuneeb Yassin

Key Points

  • This work aims to develop a data-driven framework for efficient modeling and optimization of nanosheet and forksheet transistors.
  • Generated an extensive dataset using LTSpice-based circuit simulations.
  • Developed a deep neural network as a surrogate model to learn the relationship between input parameters and transistor behavior.
  • Employed Bayesian optimization to explore the design space for high-performing configurations.
  • Achieved a coefficient of determination (R2≈0.91) and mean absolute error (MAE ≈0.024) in predictive performance on test data.
  • Demonstrated that NS devices have faster switching and lower propagation delay, while FS devices showed improved stability under specific conditions.
  • Provided a scalable methodology for surrogate modeling and design-space exploration for advanced transistor architectures.

Abstract

The continued scaling of semiconductor devices at advanced technology nodes introduces significant challenges in maintaining performance, reliability, and design efficiency. This work presents a data-driven framework for the modeling and optimization of nanosheet (NS) and forksheet (FS) transistors using deep learning and Bayesian optimization. An extensive dataset is generated through LTSpice-based circuit simulations, enabling efficient exploration of the design space while incorporating key device parameters, including channel length, channel width, supply voltage, temperature, and threshold voltage, together with variability and noise effects. A deep neural network (DNN) is developed as a surrogate model to learn the nonlinear relationship between input parameters and transistor switching behavior, achieving strong predictive performance with a coefficient of determination (R2≈0.91), mean absolute error (MAE ≈0.024), and root mean square error (RMSE ≈0.031) on unseen test data. To improve physical consistency, a bounded-output formulation is introduced to guarantee physically admissible voltage predictions, while device-level benchmarking is performed to assess agreement with expected transistor characteristics. The results demonstrate accurate modeling of transient behavior across the sampled operating conditions. Comparative analysis shows that NS devices achieve faster switching and lower propagation delay, whereas FS devices exhibit improved stability under certain conditions. Bayesian optimization is employed to efficiently explore the design space and identify high-performing transistor configurations without exhaustive simulation-based searches. The proposed framework provides a scalable and computationally efficient methodology for surrogate modeling, design-space exploration, and early-stage assessment of advanced transistor architectures.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Salman et al. (2026) studied this question.

synapsesocial.com/papers/6a338c62630953a74978dc0bhttps://doi.org/10.3390/electronics15122661
Ask AI
Helpful
Bookmark
Share
View Full Paper