Cross‐scene pavement distress detection by a novel transfer learning framework

2021 ◽  
Author(s):  
Yishun Li ◽  
Pengyu Che ◽  
Chenglong Liu ◽  
Difei Wu ◽  
Yuchuan Du
Keyword(s):  
Transfer Learning ◽  
Pavement Distress ◽  
Learning Framework ◽  
Pavement Distress Detection

scholarly journals A Model of Two Tales: Dual Transfer Learning Framework for Improved Long-tail Item Recommendation

2021 ◽  
Author(s):  
Yin Zhang ◽  
Derek Zhiyuan Cheng ◽  
Tiansheng Yao ◽  
Xinyang Yi ◽  
Lichan Hong ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Long Tail ◽  
Learning Framework

scholarly journals Cross-property deep transfer learning framework for enhanced predictive analytics on small materials data

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Vishu Gupta ◽  
Kamal Choudhary ◽  
Francesca Tavazza ◽  
Carelyn Campbell ◽  
Wei-keng Liao ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Predictive Models ◽  
Materials Science ◽  
Predictive Analytics ◽  
Large Datasets ◽  
Small Data ◽  
Widespread Application ◽  
Learning Framework ◽  
Large Databases ◽  
Physical Attributes

AbstractArtificial intelligence (AI) and machine learning (ML) have been increasingly used in materials science to build predictive models and accelerate discovery. For selected properties, availability of large databases has also facilitated application of deep learning (DL) and transfer learning (TL). However, unavailability of large datasets for a majority of properties prohibits widespread application of DL/TL. We present a cross-property deep-transfer-learning framework that leverages models trained on large datasets to build models on small datasets of different properties. We test the proposed framework on 39 computational and two experimental datasets and find that the TL models with only elemental fractions as input outperform ML/DL models trained from scratch even when they are allowed to use physical attributes as input, for 27/39 (≈ 69%) computational and both the experimental datasets. We believe that the proposed framework can be widely useful to tackle the small data challenge in applying AI/ML in materials science.

Sign in / Sign up

Export Citation Format

Share Document