Leverage electron properties to predict phonon properties via transfer learning for semiconductors

Abstract

Transfer learning greatly improves phonon prediction by leveraging knowledge hidden in bigger, even if inaccurate, electron data.

Document Details

• Document Type: Pub Defense Publication
• Publication Date: Nov 06, 2020
• Source ID: 10.1126/sciadv.abd1356

Entities

People

• Meng Jiang
• Tengfei Luo
• Zeyu Liu

Organizations

• National Science Foundation
• University of Notre Dame

Tags

Readers

• Educational Psychology
• Neural Network Machine Learning.
• Quantum Dot Semiconductor Device Photonics and Graphene Optoelectronic Materials and THz Physics.

Technology Areas

• Microelectronics