mathbbSL(n) Representation Learning: An Intrinsic Mixed-Curvature Space with Higher Curvature Capacities and Deeper Order-Aware Composition
Not on the map. Why?
Abstract (arXiv)
Mixed-curvature representation learning seeks to capture rich geometric structures that cannot be adequately modeled by a single curvature regime. Existing approaches largely rely on product manifolds, which require manually specifying how different curvature spaces are combined and separate their curvature contributions across factors. We introduce the mathbbSL(n) space, a representation geometry defined by the simple det(A)=1 constraint and a left invariant Schatten-p Finsler structure. Despite this minimal construction, mathbbSL(n) exhibits pointwise negative, zero, and positive flag curvature around a common flagpole, while its mixed-curvature and curvature-coupling capacities are asymptotically maximal relative to the intrinsic geometric upper bound. Beyond geometry, its noncommutative group structure provides inherent order sensitivity, and its non-nilpotent Lie algebra admits nonzero nested Lie brackets at arbitrary depth, enabling deep order-aware composition. Empirically, mathbbSL(n) consistently outperforms a broad range of representation manifold baselines across graph benchmarks at different scales. It reduces average distortion over the strongest baselines by 44.3% on KEGG and 40.5% on HumanCyc, and improves Hits@20 by 42.8% on OGBL-PPA. Experiments on Flickr30k-Order further support its ability to capture higher order dependencies from ordered composition. Together, these results show how a seemingly simple structural constraint can yield unexpectedly rich geometry, capacity, and composition within a unified representation space.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 43 references that are in the map. First posted on arXiv 14 Sep 2026.
Builds on
24 of 43 references.
References · 24
- 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,462
- 2021Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gaugescited by 1,131
- 2017Poincar'e Embeddings for Learning Hierarchical Representationscited by 674
- 2019Hyperbolic Graph Convolutional Neural Networkscited by 400
- 2020Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Datacited by 248
- 2020Low-Dimensional Hyperbolic Knowledge Graph Embeddingscited by 195
- 2019Constant Curvature Graph Convolutional Networkscited by 106
- 2020LieTransformer: Equivariant self-attention for Lie Groupscited by 100
- 2019Mixed-curvature Variational Autoencoderscited by 85
- 2021Automatic Symmetry Discovery with Lie Algebra Convolutional Networkcited by 82
- 2016Building Deep Networks on Grassmann Manifoldscited by 58
- 2016Deep Learning on Lie Groups for Skeleton-based Action Recognitioncited by 42
12 more
- 2021A Self-supervised Mixed-curvature Graph Neural Networkcited by 32
- 2021Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approachcited by 23
- 2024GraphMoRE: Mitigating Topological Heterogeneity via Mixture of Riemannian Expertscited by 23
- 2023Lie Group Decompositions for Equivariant Neural Networkscited by 12
- 2023Modeling Graphs Beyond Hyperbolic: Graph Neural Networks in Symmetric Positive Definite Matricescited by 10
- 2025Finsler Multi-Dimensional Scaling: Manifold Learning for Asymmetric Dimensionality Reduction and Embeddingcited by 7
- 2023Lie Neurons: Adjoint-Equivariant Neural Networks for Semisimple Lie Algebrascited by 6
- 2023Improving Heterogeneous Graph Learning with Weighted Mixed-Curvature Product Manifoldcited by 4
- 2024Product Manifold Representations for Learning on Biological Pathwayscited by 4
- 2025Equivariant Neural Networks for General Linear Symmetries on Lie Algebrascited by 4
- 2023Learning Polynomial Problems with SL(2,ℝ) Equivariancecited by 2
- 2026Finsler Geometry, Graph Neural Networks, and Youcited by 1
Built on by
Not cited yet.