The Huber Mean: A Robust Way to Average Data on Curved Spaces (IMAGE)

Pusan National University

The Huber Mean: A Robust Way to Average Data on Curved Spaces

Caption

The newly developed Huber mean provides a more stable and reliable way to compute averages for data lying on curved geometric spaces, or Riemannian manifolds. By combining the strengths of least-squares (L₂) and least-absolute-deviation (L₁) methods, the Huber mean resists distortion from outliers while maintaining efficiency, improving data analysis in fields such as medical imaging, robotics, and artificial intelligence.

Credit

Pusan National University

Usage Restrictions

Cannot be reused without permission

License

Original content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.