# Hyperbolic Coarse-to-Fine Few-Shot Class-Incremental Learning

arXiv: https://arxiv.org/abs/2509.18504 · HTML: https://mlsanity.com/paper/2509.18504/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 30 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Builds on (most cited first, 24 of 24)

- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning](https://mlsanity.com/paper/1711.05769/) (2017)
- [Meta-Learning for Semi-Supervised Few-Shot Classification](https://mlsanity.com/paper/1803.00676/) (2018)
- [Hyperbolic Neural Networks](https://mlsanity.com/paper/1805.09112/) (2018)
- [Meta-Transfer Learning for Few-Shot Learning](https://mlsanity.com/paper/1812.02391/) (2018)
- [Learning without Memorizing](https://mlsanity.com/paper/1811.08051/) (2018)
- [Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learning](https://mlsanity.com/paper/1903.12290/) (2019)
- [Continuous Hierarchical Representations with Poincar'e Variational Auto-Encoders](https://mlsanity.com/paper/1901.06033/) (2019)
- [Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning](https://mlsanity.com/paper/2102.03983/) (2021)
- [Generating Classification Weights with GNN Denoising Autoencoders for Few-Shot Learning](https://mlsanity.com/paper/1905.01102/) (2019)
- [Hyperbolic Contrastive Learning for Visual Representations beyond Objects](https://mlsanity.com/paper/2212.00653/) (2022)
- [Differentiating through the Fr'echet Mean](https://mlsanity.com/paper/2003.00335/) (2020)
- [The Numerical Stability of Hyperbolic Representation Learning](https://mlsanity.com/paper/2211.00181/) (2022)
- [Fine-grained Angular Contrastive Learning with Coarse Labels](https://mlsanity.com/paper/2012.03515/) (2020)
- [Generative Adversarial Residual Pairwise Networks for One Shot Learning](https://mlsanity.com/paper/1703.08033/) (2017)
- [Coarse-To-Fine Incremental Few-Shot Learning](https://mlsanity.com/paper/2111.14806/) (2021)
- [Hyperbolic Space with Hierarchical Margin Boosts Fine-Grained Learning from Coarse Labels](https://mlsanity.com/paper/2311.11019/) (2023)
- [Towards Few-Shot Learning in the Open World: A Review and Beyond](https://mlsanity.com/paper/2408.09722/) (2024)
- [A Capsule Network for Hierarchical Multi-Label Image Classification](https://mlsanity.com/paper/2209.05723/) (2022)
- [Coarse-to-Fine: Learning Compact Discriminative Representation for Single-Stage Image Retrieval](https://mlsanity.com/paper/2308.04008/) (2023)
- [Learning from Adversarial Features for Few-Shot Classification](https://mlsanity.com/paper/1903.10225/) (2019)
- [OpenEarthSensing: Large-Scale Fine-Grained Benchmark for Open-World Remote Sensing](https://mlsanity.com/paper/2502.20668/) (2025)

## Abstract (arXiv)

In the field of machine learning, hyperbolic space demonstrates superior representation capabilities for hierarchical data compared to conventional Euclidean space. This work focuses on the Coarse-To-Fine Few-Shot Class-Incremental Learning (C2FSCIL) task. Our study follows the Knowe approach, which contrastively learns coarse class labels and subsequently normalizes and freezes the classifier weights of learned fine classes in the embedding space. To better interpret the "coarse-to-fine" paradigm, we propose embedding the feature extractor into hyperbolic space. Specifically, we employ the Poincar'e ball model of hyperbolic space, enabling the feature extractor to transform input images into feature vectors within the Poincar'e ball instead of Euclidean space. We further introduce hyperbolic contrastive loss and hyperbolic fully-connected layers to facilitate model optimization and classification in hyperbolic space. Additionally, to enhance performance under few-shot conditions, we implement maximum entropy distribution in hyperbolic space to estimate the probability distribution of fine-class feature vectors. This allows generation of augmented features from the distribution to mitigate overfitting during training with limited samples. Experiments on C2FSCIL benchmarks show that our method effectively improves both coarse and fine class accuracies.