hyperbolic space
PulseAugur coverage of hyperbolic space — every cluster mentioning hyperbolic space across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New HyperFlow method enables incremental semantic change detection in remote sensing
Researchers have developed HyperFlow, a novel method for class-incremental semantic change detection in remote sensing. This approach addresses the limitation of existing methods that assume a fixed set of change catego…
-
New EH-Fusion framework tackles Euclidean crowding in hyperbolic graph learning
Researchers have developed a new framework called EH-Fusion to address Euclidean crowding in hyperbolic graph representation learning. This framework provides a geometric quantification of Euclidean crowding and a theor…
-
Hyperbolic space boosts patent prior-art search over Euclidean, study finds
A new research paper explores the use of embedding geometries and clustering algorithms for unsupervised patent classification, finding that hyperbolic space offers a significant advantage over Euclidean space. The stud…
-
New Gromov-Wasserstein method aligns distributions across diverse geometric spaces
Researchers have developed a new divergence metric called Constant-Curvature Sliced Gromov-Wasserstein (CCSGW) to align probability distributions across heterogeneous constant-curvature spaces, such as hyperbolic and sp…
-
New Research Explores Implicit Bias in Hyperbolic Representation Learning
This paper delves into the implicit bias of hyperbolic representation learning for multiclass data, analyzing it through the lens of Busemann risk. The research decomposes the distance to class prototypes in hyperbolic …
-
New HCFormer Architecture Uses Hyperbolic Clustering for Interpretable Vision Models
Researchers have developed HCFormer, a new vision backbone architecture that utilizes hyperbolic hierarchical clustering for visual representation learning. This approach, named ClusterMixer, offers a more interpretable…
-
HyperANFIS uses hyperbolic geometry to boost fuzzy inference systems
Researchers have introduced HyperANFIS, a novel extension of Adaptive Neuro-Fuzzy Inference Systems (ANFIS) that leverages hyperbolic geometry. Traditional ANFIS models operate in Euclidean space, which can limit their …
-
New AI framework H2AL uses hyperbolic space for medical image segmentation
Researchers have introduced H2AL, a novel framework for few-shot medical image segmentation that utilizes hyperbolic space to better model anatomical hierarchies. This approach, detailed in a recent arXiv paper, aims to…
-
New MGFA statistical model enhances manifold data clustering
Researchers have introduced Mixtures of Geodesic Factor Analyzers (MGFA), a novel statistical model designed for clustering manifold-valued data. MGFA offers enhanced expressiveness compared to existing methods by incor…
-
New PINN framework detailed for geometric analysis problems
This paper, a companion to a previous work, details a machine learning framework using physics-informed neural networks (PINNs) for constructing minimal discs in hyperbolic space. It focuses on methodological aspects, e…
-
New framework unifies Riemannian deep learning modules and geometries
A new thesis proposes a unified framework for Riemannian deep learning, addressing challenges with manifold-valued representations. The work introduces reusable neural modules, manifold-specific network architectures, a…
-
PoLAR method enhances robot policy learning with structured latent actions
Researchers have developed PoLAR, a novel approach to robot policy learning that utilizes a geometrically structured latent action representation within hyperbolic space. This method disentangles transition extent from …
-
New HyRAG framework boosts graph model generalization
Researchers have developed a new framework called Hyperbolic Retrieval-Augmented Generation (HyRAG) to improve the generalization capabilities of Graph Foundation Models (GFMs). Existing RAG methods struggle with the ge…