Moderately Supervised Learning: Definition, Framework and Generality

Supervised learning (SL) has achieved remarkable success in numerous artificial intelligence (AI) applications. In the current literature, by referring to the properties of the ground-truth labels prepared for the training data set, SL is roughly categorized as fully supervised learning (FSL) and weakly supervised learning (WSL). FSL concerns the situation where the training data set is assigned with ideal ground-truth labels, while WSL concerns the situation where the training data set is assigned with non-ideal ground-truth labels. However, solutions for various FSL tasks have shown that the given ground-truth labels are not always learnable, and the target transformation from the given ground-truth labels to learnable targets can significantly affect the performance of the final FSL solutions. The roughness of the FSL category conceals some details that are critical to building the appropriate solutions for some specific FSL tasks. In this paper, taking into consideration the properties of the target transformation from the given ground-truth labels to learnable targets, we firstly categorize FSL into three narrower sub-types and then focus on the sub-type moderately supervised learning (MSL) that concerns the situation where the given ground-truth labels are ideal, but due to the simplicity in annotation of the given ground-truth labels, careful designs are required to transform the given ground-truth labels into learnable targets. From the perspectives of the definition, framework and generality, we comprehensively illustrate MSL to reveal what details are concealed by the roughness of the FSL category. At the meantime, via presenting the definition, framework and generality of MSL, this paper as well establishes a tutorial for AI application engineers to refer to viewing a problem to be solved from the mathematicians' vision.