APPLYING SELF-CONFIDENCE IN MULTI-LABEL CLASSIFICATION TO MODEL TRAINING

• Main Authors: Thyagharajan, Anirud, Laddha, Prashant, Ummenhofer, Benjamin, Omer, Om Ji
• Format: Patent
• Language: English
• Published: 17.03.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; PHYSICS

A computer model is trained to classify regions of a space (e.g., a pixel of an image or a voxel of a point cloud) according to a multi-label classification. To improve the model's accuracy, the model's self-confidence is determined with respect to its own predictions of regions in a train...