In safety-critical applications, microcontrollers must meet performance standards, including the maximum operating frequency (Fmax). ML models can estimate Fmax using data from on-chip ring oscillators (ROs). However, when new products are introduced, existing ML models may no longer be suitable and may require updating. Training a new model is challenging due to limited data availability and time needed to acquire Fmax. But data from legacy products, along with pre-trained models, may still be available. We propose using deep-learning models trained on a specific MCU product as feature extractors for new devices, to address the scarcity of labeled data, in a Transfer Learning fashion. Experimental results show that these models can extract useful general features for performance prediction even from new products. As a result, they achieve better performance with significantly less labeled data compared to traditional shallow learning approaches

Transfer Learning in MCU Performance Screening / Bellarmino, Nicolo; Cantoro, Riccardo; Huch, Martin; Kilian, Tobias; Schlichtmann, Ulf; Squillero, Giovanni. - ELETTRONICO. - (In corso di stampa). (Intervento presentato al convegno IEEE International Test Conference (ITC 2023) tenutosi a Anaheim, CA 92802, Stati Uniti nel 8-13 Ottobre 2023).

Transfer Learning in MCU Performance Screening

Bellarmino, Nicolo;Cantoro, Riccardo;Squillero, Giovanni
In corso di stampa

Abstract

In safety-critical applications, microcontrollers must meet performance standards, including the maximum operating frequency (Fmax). ML models can estimate Fmax using data from on-chip ring oscillators (ROs). However, when new products are introduced, existing ML models may no longer be suitable and may require updating. Training a new model is challenging due to limited data availability and time needed to acquire Fmax. But data from legacy products, along with pre-trained models, may still be available. We propose using deep-learning models trained on a specific MCU product as feature extractors for new devices, to address the scarcity of labeled data, in a Transfer Learning fashion. Experimental results show that these models can extract useful general features for performance prediction even from new products. As a result, they achieve better performance with significantly less labeled data compared to traditional shallow learning approaches
In corso di stampa
File in questo prodotto:
File Dimensione Formato  
2023_XXX_Transfer_Learning_POSTER (5).pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 119.55 kB
Formato Adobe PDF
119.55 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2981875