Skip to main content
Forsíða
Forsíða

Few-Shot Remote Sensing Classification with Meta-Learning: Challenges and Opportunities

Few-Shot Remote Sensing Classification with Meta-Learning: Challenges and Opportunities - Available at University of Iceland
When 
Thu, 28/04/2022 - 12:30 to 13:00
Where 
Further information 
Free admission

Surbhi Sharma, PhD student in Computational Engineering at the University of Iceland and the Jülich Supercomputing Centre (JSC), Germany gives the lecture Few-Shot Remote Sensing Classification with Meta-Learning: Challenges and Opportunities.

Event on Zoom:
https://eu01web.zoom.us/my/crs.talks.2022
Meeting ID: 354 191 1354

Abstract:

Obtaining high generalization and classification accuracies with limited availability of annotated data and from data having diverse distributions is often challenging in the field of remote sensing.
Few-shot learning with the help of meta-learning can tackle meta-problems where learning algorithm learns to learn new tasks or classes from limited training data, thereby decreasing the learning time with improved generalization.

 

 Surbhi Sharma

Few-Shot Remote Sensing Classification with Meta-Learning: Challenges and Opportunities