Hierarchical Class-Based Curriculum Loss
This addresses the issue of flat label assumptions in classification for domains with hierarchical data, offering improved accuracy and interpretability, though it is incremental as it builds on existing hierarchical methods.
The paper tackles the problem of classification with hierarchical label dependencies by proposing a hierarchical curriculum loss that satisfies constraints and assigns non-uniform weights to labels, showing it is a tighter bound of 0-1 loss and significantly outperforms baselines on real-world image datasets.
Classification algorithms in machine learning often assume a flat label space. However, most real world data have dependencies between the labels, which can often be captured by using a hierarchy. Utilizing this relation can help develop a model capable of satisfying the dependencies and improving model accuracy and interpretability. Further, as different levels in the hierarchy correspond to different granularities, penalizing each label equally can be detrimental to model learning. In this paper, we propose a loss function, hierarchical curriculum loss, with two properties: (i) satisfy hierarchical constraints present in the label space, and (ii) provide non-uniform weights to labels based on their levels in the hierarchy, learned implicitly by the training paradigm. We theoretically show that the proposed loss function is a tighter bound of 0-1 loss compared to any other loss satisfying the hierarchical constraints. We test our loss function on real world image data sets, and show that it significantly substantially outperforms multiple baselines.