A Transfer Learning Approach to Expedite Training of Artificial Neural Networks for Variability-Aware Signal Integrity Analysis of MWCNT Interconnects

Guglani, Surila; Dimple, Km; DasGupta, Anirvan; Sharma, Rohit; Kaushik, Brajesh Kumar; Roy, Sourajeet

doi:10.1109/epeps53828.2022.9947162

Cited by 6 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural networks are also very data hungry, and the related approaches require a massive amount of simulations of the actual model to achieve a reasonable training accuracy. This issue can be mitigated by leveraging suitable knowledge-based networks [22] or multiple neural networks trained at different levels of fidelity [23], when applicable. Not surprisingly, neural networks turn out to be effective mainly for those tasks for which huge datasets are available, e.g., image processing.…”

Section: Introductionmentioning

confidence: 99%

Conservative Gaussian Process Models for Uncertainty Quantification and Bayesian Optimization in Signal Integrity Applications

Manfredi

2024

IEEE Trans. Compon., Packag. Manufact. Technol.

View full text Add to dashboard Cite

Surrogate modeling is being increasingly adopted in signal and power integrity analysis to assist design exploration, optimization, and uncertainty quantification tasks. In this scenario, machine learning methods are attracting an ever-growing interest over alternative and well-consolidated techniques due to their data-driven nature. However, an open issue is to properly assess the trustworthiness of predictions when generalizing beyond training data. Among various machine learning tools, Gaussian process regression (GPR) has the notable feature of providing an estimate of the prediction uncertainty (or confidence) due to the lack of data. Nevertheless, the uncertainty introduced by the estimation of the Gaussian process parameters, which is part of the training process, is typically not accounted for. In this paper, we introduce improved GPR formulations that take into account the additional uncertainty related to the estimation of (some of) the Gaussian process parameters, thereby providing a more accurate estimate of the actual prediction confidence. Furthermore, the advocated framework is extended to uncertainty quantification and Bayesian optimization settings. The technique is applied to two test cases concerning the analysis of crosstalk in a transmission line network and of the frequencydomain response of a microstrip line with a ground plane discontinuity.

show abstract

Section: Introductionmentioning

confidence: 99%