Transfer Learning for Minimum Operating Voltage Prediction in Advanced Technology Nodes: Leveraging Legacy Data and Silicon Odometer Sensing
This work addresses the problem of chip performance prediction for semiconductor manufacturers, enabling better energy efficiency and reliability, though it is incremental as it builds on existing transfer learning methods with new sensor integration.
The paper tackled the challenge of predicting minimum operating voltage (V_min) at advanced 5nm technology nodes by proposing a transfer learning framework that leverages legacy data from 16nm nodes and integrates silicon odometer sensor data, achieving significantly improved prediction accuracy.
Accurate prediction of chip performance is critical for ensuring energy efficiency and reliability in semiconductor manufacturing. However, developing minimum operating voltage ($V_{min}$) prediction models at advanced technology nodes is challenging due to limited training data and the complex relationship between process variations and $V_{min}$. To address these issues, we propose a novel transfer learning framework that leverages abundant legacy data from the 16nm technology node to enable accurate $V_{min}$ prediction at the advanced 5nm node. A key innovation of our approach is the integration of input features derived from on-chip silicon odometer sensor data, which provide fine-grained characterization of localized process variations -- an essential factor at the 5nm node -- resulting in significantly improved prediction accuracy.