On Transfer in Classification: How Well do Subsets of Classes Generalize?
This work provides foundational insights into transfer mechanics for researchers in machine learning, though it is incremental as it builds on existing concepts of generalization and transfer learning.
The paper tackles the problem of understanding how classification models generalize from trained subsets of classes to unseen ones, establishing a theoretical framework using partially ordered sets to represent transferability and exploring its practical application in predicting optimal subsets and few-shot learning.
In classification, it is usual to observe that models trained on a given set of classes can generalize to previously unseen ones, suggesting the ability to learn beyond the initial task. This ability is often leveraged in the context of transfer learning where a pretrained model can be used to process new classes, with or without fine tuning. Surprisingly, there are a few papers looking at the theoretical roots beyond this phenomenon. In this work, we are interested in laying the foundations of such a theoretical framework for transferability between sets of classes. Namely, we establish a partially ordered set of subsets of classes. This tool allows to represent which subset of classes can generalize to others. In a more practical setting, we explore the ability of our framework to predict which subset of classes can lead to the best performance when testing on all of them. We also explore few-shot learning, where transfer is the golden standard. Our work contributes to better understanding of transfer mechanics and model generalization.