Achievement
Loss function design in machine learning
Project
Vision and Learning in Humans and Machines
University
University of California at San Diego
(La Jolla, CA)
Research Achievements
Loss function design in machine learning
IGERT fellow Hamed Masnadi-Shirazi, in collaboration with Vijay Mahadevan and IGERT affiliated faculty member Nuno Vasconcelos, has applied the principles of classifier loss function design from machine learning (previously published as a 2008 NIPS paper) to the problem of object tracking and scene classification in computer vision. The insights gained from the subject of loss function design in machine learning has allowed for the design of outlier resistant and robust classification algorithms that produce state of the art results in object tracking and scene classification in computer vision. These results have been accepted for publication in CVPR 2010.
- “Research Achievements”
- Achievements for this Project