paper-with-me

홈 › Papers

A Coding-Theoretic Analysis of Hyperspherical Prototypical Learning Geometry

2024-07-10 · Martin Lindström, Borja Rodríguez-Gálvez, Ragnar Thobaben, Mikael Skoglund

Hyperspherical Prototypical Learning (HPL) is a supervised approach to representation learning that designs class prototypes on the unit hypersphere. The prototypes bias the representations to class separation in a scale invariant and known geometry. Previous approaches to HPL have either of the following shortcomings: (i) they follow an unprincipled optimisation procedure; or (ii) they are theoretically sound, but are constrained to only one possible latent dimension. In this paper, we address both shortcomings. To address (i), we present a principled optimisation procedure whose solution we show is optimal. To address (ii), we construct well-separated prototypes in a wide range of dimensions using linear block codes. Additionally, we give a full characterisation of the optimal prototype placement in terms of achievable and converse bounds, showing that our proposed methods are near-optimal.

📄 PDF Abstract BibTeX arXiv:2407.07664

Code (1)

martinlindstrom/coding_theoretic_hpl 공식 구현 pytorch

Tasks

Representation Learning

Similar Papers 제목 키워드 기반

Harmonizing Geometry and Uncertainty: Diffusion with Hyperspheres

2025-06-12 · Muskan Dosi, Chiranjeev Chiranjeev, Kartik Thakral, Mayank Vatsa 외

Do contemporary diffusion models preserve the class geometry of hyperspherical data? Standard diffusion models rely on isotropic Gaussian noise in the forward process, inherently favoring Euclidean spaces. However, many …

HYPO: Hyperspherical Out-of-Distribution Generalization

2024-02-12 · Haoyue Bai, Yifei Ming, Julian Katz-Samuels, Yixuan Li

Out-of-distribution (OOD) generalization is critical for machine learning models deployed in the real world. However, achieving this can be fundamentally challenging, as it requires the ability to learn invariant feature…

Out-of-Distribution Generalization

Self-Supervised Representation Learning via Hyperspherical Density Shaping

2026-04-27 · Esteban Rodríguez-Betancourt, Edgar Casasola-Murillo arxiv

Modern self-supervised representation learning methods often relies on empirical heuristics that are not theoretically grounded. In this study we propose HyDeS, a theoretically grounded method based on multi-view mutual …

Self-Supervised LearningRepresentation Learning

Hyperspherical Latents Improve Continuous-Token Autoregressive Generation

2025-09-29 · Guolin Ke, Hui Xue arxiv

Autoregressive (AR) models are promising for image generation, yet continuous-token AR variants often trail latent diffusion and masked-generation models. The core issue is heterogeneous variance in VAE latents, which is…

Image Generation

Geometry-Preserving Aggregation for Mixture-of-Experts Embedding Models

2026-02-15 · Sajjad Kachuee, Mohammad Sharifkhani arxiv

Mixture-of-Experts (MoE) embedding models combine expert outputs using weighted linear summation, implicitly assuming a linear subspace structure in the embedding space. This assumption is shown to be inconsistent with t…

Semantic Similarity