# Hierarchical Class-Based Curriculum Loss

arXiv: https://arxiv.org/abs/2006.03629 · HTML: https://mlsanity.com/paper/2006.03629/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jun 2020.

## Built on by (most cited first, 5 of 5)

- [Skin Lesion Recognition with Class-Hierarchy Regularized Hyperbolic Embeddings](https://mlsanity.com/paper/2209.05842/) (2022)
- [United We Learn Better: Harvesting Learning Improvements From Class Hierarchies Across Tasks](https://mlsanity.com/paper/2107.13627/) (2021)
- [Towards Understanding the Effect of Pretraining Label Granularity](https://mlsanity.com/paper/2303.16887/) (2023)
- [Tree semantic segmentation from aerial image time series](https://mlsanity.com/paper/2407.13102/) (2024)
- [hYOLO Model: Enhancing Object Classification with Hierarchical Context in YOLOv8](https://mlsanity.com/paper/2510.23278/) (2025)

## Abstract (arXiv)

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.