Methods and equipment for training machine learning models

Abstract

This morning's discussion covered general methods and equipment for training machine learning models. The specific aspects of the invention propose machine learning techniques using base separation, including: receiving the first runtime record, where the first runtime record contains signal data Rl·' collected in physical space; processing the first runtime record using multiple base machine learning (ML) models to generate multiple inferences; synthesizing multiple inferences to generate predictions including multiple coordinates; and output a prediction, which includes coordinates indicating the position of the physical element in physical space.