Combined Hyperbolic and Euclidean Soft Triple Loss Beyond the Single Space Deep Metric Learning

• URL: http://arxiv.org/abs/2510.05643v1
• Date: Tue, 07 Oct 2025 07:42:49 GMT
• Title: Combined Hyperbolic and Euclidean Soft Triple Loss Beyond the Single Space Deep Metric Learning
• Abstract summary: We find that the combination of hyperbolic and Euclidean spaces improves DML accuracy and learning stability for both spaces.<n>We evaluate the CHEST loss on four benchmark datasets, achieving a new state-of-the-art performance.
• Score: 0.0
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Deep metric learning (DML) aims to learn a neural network mapping data to an embedding space, which can represent semantic similarity between data points. Hyperbolic space is attractive for DML since it can represent richer structures, such as tree structures. DML in hyperbolic space is based on pair-based loss or unsupervised regularization loss. On the other hand, supervised proxy-based losses in hyperbolic space have not been reported yet due to some issues in applying proxy-based losses in a hyperbolic space. However, proxy-based losses are attractive for large-scale datasets since they have less training complexity. To address these, this paper proposes the Combined Hyperbolic and Euclidean Soft Triple (CHEST) loss. CHEST loss is composed of the proxy-based losses in hyperbolic and Euclidean spaces and the regularization loss based on hyperbolic hierarchical clustering. We find that the combination of hyperbolic and Euclidean spaces improves DML accuracy and learning stability for both spaces. Finally, we evaluate the CHEST loss on four benchmark datasets, achieving a new state-of-the-art performance.

Related papers

• Hyperbolic Enhanced Representation Learning for Incomplete Multi-view Clustering [57.38215918201251]
  We propose HERL, a Hyperbolic Enhanced Representation Learning framework for Incomplete Multi-View Clustering.<n> operating within the Poincaré ball, HERL constructs a structure-aware latent space to enhance representation learning.<n>We show that HERL consistently outperforms state-of-the-art approaches.
  arXiv  Detail & Related papers  (2026-04-18T10:50:46Z)
• Hyperbolic Hierarchical Alignment Reasoning Network for Text-3D Retrieval [38.14508593270809]
  We introduce the Hyperbolic Hierarchical Alignment Reasoning Network (H$2$ARN) for text-3D retrieval.<n>H$2$ARN embeds both text and 3D data in a Lorentz-model hyperbolic space.<n>We also release the expanded T3DR-HIT v2 benchmark, which contains 8,935 text-to-3D pairs, 2.6 times the original size.
  arXiv  Detail & Related papers  (2025-11-14T07:59:57Z)
• Geometry-aware Distance Measure for Diverse Hierarchical Structures in Hyperbolic Spaces [48.948334221681684]
  We propose a geometry-aware distance measure in hyperbolic spaces, which dynamically adapts to varying hierarchical structures.<n>Our approach consistently outperforms learning methods that use fixed distance measures.<n>Visualization shows clearer class boundaries and improved prototype separation in hyperbolic spaces.
  arXiv  Detail & Related papers  (2025-06-23T11:43:39Z)
• Understanding and Mitigating Hyperbolic Dimensional Collapse in Graph Contrastive Learning [70.0681902472251]
  We propose a novel contrastive learning framework to learn high-quality graph embeddings in hyperbolic space.<n>Specifically, we design the alignment metric that effectively captures the hierarchical data-invariant information.<n>We show that in the hyperbolic space one has to address the leaf- and height-level uniformity related to properties of trees.
  arXiv  Detail & Related papers  (2023-10-27T15:31:42Z)
• HyP$^2$ Loss: Beyond Hypersphere Metric Space for Multi-label Image Retrieval [20.53316810731414]
  We propose a novel metric learning framework with Hybrid Proxy-Pair Loss (HyP$2$ Loss) The proposed HyP$2$ Loss focuses on optimizing the hypersphere space by learnable proxies and excavating data-to-data correlations of irrelevant pairs.
  arXiv  Detail & Related papers  (2022-08-14T15:06:27Z)
• Towards Scalable Hyperbolic Neural Networks using Taylor Series Approximations [10.056167107654089]
  Hyperbolic networks have shown prominent improvements over their Euclidean counterparts in several areas involving hierarchical datasets. Their adoption in practice remains restricted due to (i) non-scalability on accelerated deep learning hardware, (ii) vanishing due to the closure of hyperbolic space, and (iii) information loss. We propose the approximation of hyperbolic operators using Taylor series expansions, which allows us to reformulate the tangent gradients of hyperbolic functions into their equivariants.
  arXiv  Detail & Related papers  (2022-06-07T22:31:17Z)
• Geometric Graph Representation Learning via Maximizing Rate Reduction [73.6044873825311]
  Learning node representations benefits various downstream tasks in graph analysis such as community detection and node classification. We propose Geometric Graph Representation Learning (G2R) to learn node representations in an unsupervised manner. G2R maps nodes in distinct groups into different subspaces, while each subspace is compact and different subspaces are dispersed.
  arXiv  Detail & Related papers  (2022-02-13T07:46:24Z)
• The KFIoU Loss for Rotated Object Detection [115.334070064346]
  In this paper, we argue that one effective alternative is to devise an approximate loss who can achieve trend-level alignment with SkewIoU loss. Specifically, we model the objects as Gaussian distribution and adopt Kalman filter to inherently mimic the mechanism of SkewIoU. The resulting new loss called KFIoU is easier to implement and works better compared with exact SkewIoU.
  arXiv  Detail & Related papers  (2022-01-29T10:54:57Z)
• CO-SNE: Dimensionality Reduction and Visualization for Hyperbolic Data [10.618060176686916]
  We propose CO-SNE, extending the Euclidean space visualization tool, t-SNE, to hyperbolic space. Unlike Euclidean space, hyperbolic space is inhomogeneous: a volume could contain a lot more points at a location far from the origin. We apply CO-SNE to high-dimensional hyperbolic biological data as well as unsupervisedly learned hyperbolic representations.
  arXiv  Detail & Related papers  (2021-11-30T00:21:47Z)
• Multi Proxy Anchor Family Loss for Several Types of Gradients [0.0]
  This paper proposes three multi-proxies anchor (MPA) family losses and a normalized discounted cumulative gain (nDCG@k) metric. MPA-family losses improve the training capacity of a neural network owing to solving the gradient problem. We demonstrate the effectiveness of MPA-family losses, and MPA-family losses achieves higher accuracy on two datasets for fine-grained images.
  arXiv  Detail & Related papers  (2021-10-08T09:34:38Z)
• Robust Large-Margin Learning in Hyperbolic Space [64.42251583239347]
  We present the first theoretical guarantees for learning a classifier in hyperbolic rather than Euclidean space. We provide an algorithm to efficiently learn a large-margin hyperplane, relying on the careful injection of adversarial examples. We prove that for hierarchical data that embeds well into hyperbolic space, the low embedding dimension ensures superior guarantees.
  arXiv  Detail & Related papers  (2020-04-11T19:11:30Z)
• clDice -- A Novel Topology-Preserving Loss Function for Tubular Structure Segmentation [57.20783326661043]
  We introduce a novel similarity measure termed centerlineDice (short clDice) We theoretically prove that clDice guarantees topology preservation up to homotopy equivalence for binary 2D and 3D segmentation. We benchmark the soft-clDice loss on five public datasets, including vessels, roads and neurons (2D and 3D)
  arXiv  Detail & Related papers  (2020-03-16T16:27:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.