Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,051 papers · 148 categories

Trend · papers per month

2.5%5.0%7.5%10.0% · Jun 199419922001200920182026
48 results for k-dimensional GNNs

Graph neural networks are shown to be as powerful as a graph isomorphism heuristic, leading to a new approach for higher-order graph structures.

problem Understanding and distinguishing non-isomorphic graphs and their higher-order structures.
method Relating GNNs to the Weisfeiler-Leman heuristic and proposing kk-dimensional GNNs.
result GNNs have the same expressiveness as the Weisfeiler-Leman heuristic in distinguishing graphs and their higher-order structures.

Introduces fractional k-dimensional measure bridging fractional length and area.

problem Defining fractional measures for dimensions between 0 and n-1.
method Introduces a parameterized fractional measure σσ that converges to Hausdorff measure.
result Fractional measure converges to Hausdorff measure with a known constant factor.

Study growth of systoles in arithmetic manifolds, focusing on kk-dimensional cases.

problem Growth of systoles in arithmetic nn-manifolds along congruence coverings.
method Analyzes growth of kk-dimensional systoles in arithmetic nn-manifolds, proving polylogarithmic and constant power bounds.
result Growth of systoles for k=rk = r oscillates between a power of a logarithm and a power function of the degree of the covering.

This paper examines the category C^k_{d,n} whose morphisms are d-dimensional smooth manifolds that are properly embedded in the product of a k-dimensional cube with an (d+n-k)-dimensional Euclidean space. There are k directions to compose k-dimensional cubes, so C^k_{d,n} is a (strict) k-tuple category. The geometric r…

2011-02-21abs ↗pdf ↗

The paper studies the topology of hyperspaces of k-dimensional convex sets.

problem Topology of hyperspaces of k-dimensional closed convex sets.
method Proved that hyperspaces are Hilbert cube manifolds with fiber bundle structure over Grassmann manifold.
result Fiber of Kk,bnK_{k,b}^n is homeomorphic with Rk(k+1)+2n2imesQ\mathbb R^{\frac{k(k+1)+2n}{2}} imes Q.

In 1972, Marcel Berger defined a metric invariant that captures the `size' of k-dimensional homology of a Riemannian manifold. This invariant came to be called the k-dimensional SYSTOLE. He asked if the systoles can be constrained by the volume, in the spirit of the 1949 theorem of C. Loewner. We construct metrics, ins…

1997-07-03abs ↗pdf ↗

Eigen-GNN enhances GNNs by preserving graph structures.

problem Existing shallow GNNs fail to effectively preserve graph structures.
method Integrates eigenspace of graph structures into GNNs as a dimensionality reduction module.
result Eigen-GNN boosts GNNs' ability to preserve graph structures without increasing depth.

A manifold is locally \emph{kk-fold symmetric}, if for any point and any kk-dimensional vector subspace tangent to this point there exists a local isometry such that this point is a fixed point and the differential of the isometry restricted to that kk-dimensional vector subspace is minus the identity. We show that …

2016-07-19abs ↗pdf ↗

In hyperbolic space HnH^n we set a geodesic ball of radius ρρ. Consider a kk dimensional minimal submanifold passing through the origin of the geodesic ball with boundary lies on the boundary of that geodesic ball. We prove that its area is no less than the totally geodesic kk dimensional submanifold passing through…

2016-12-08abs ↗pdf ↗

Let ΣΣ be a kk-dimensional minimal submanifold in the nn-dimensional unit ball BnB^n which passes through a point yBny \in B^n and satisfies ΣBn\partial Σ\subset \partial B^n. We show that the kk-dimensional area of ΣΣ is bounded from below by Bk(1y2)k2|B^k| \, (1-|y|^2)^{\frac{k}{2}}. This settles a question left open by …

2016-07-15abs ↗pdf ↗

The paper deals with amoebas of kk-dimensional algebraic varieties in the algebraic complex torus of dimension n2kn\geq 2k. First, we show that the area of complex algebraic curve amoebas is finite. Moreover, we give an estimate of this area in the rational curve case in terms of the degree of the rational parametrizat…

2011-01-25abs ↗pdf ↗

We prove that Dranishnikov's kk-dimensional resolution dk ⁣:μkQd_k\colon μ^k\to Q is a UVn1^{n-1}-divider of Chigogidze's kk-dimensional resolution ckc_k. This fact implies that dk1d_k^{-1} preserves ZZ-sets. A further development of the concept of UVn1^{n-1}-dividers permits us to find sufficient conditions for $d_k^{-1}(…

2008-03-28abs ↗pdf ↗

SHAKE-GNN scales GNNs for large graphs with multi-scale representations.

problem Scaling Graph Neural Networks (GNNs) to large graphs.
method SHAKE-GNN uses a hierarchy of Kirchhoff Forests for stochastic multi-resolution graph decompositions.
result SHAKE-GNN achieves competitive performance on large-scale graph classification benchmarks.

LC-GNN improves GNNs for node classification by incorporating label consistency.

problem Limited performance of GNNs due to label consistency assumption not always holding.
method LC-GNN uses node pairs with the same label but unconnected to expand GNN's receptive field.
result LC-GNN outperforms traditional GNNs in semi-supervised node classification.

New insights into GNN optimization reveal skip connections and depth accelerate training.

problem Understanding and optimizing the training of Graph Neural Networks (GNNs).
method Analysis of gradient dynamics in linearized GNNs and empirical validation.
result GNNs are implicitly accelerated by skip connections, more depth, and good label distribution during training.

Hardness proven for embedding simplicial complexes in R^d, especially for k-dimensional ones.

problem Recognizing almost embeddability of k-dimensional complexes in R^d.
method NP-hardness proof using configuration spaces and preimage cycle properties.
result Embedding obstruction is incomplete for k-dimensional complexes in R^d.

DefNet defends GNNs against adversarial attacks by identifying and mitigating vulnerabilities.

problem Vulnerability of GNNs to adversarial attacks.
method Investigates latent vulnerabilities in GNN layers, proposes dual-stage aggregation and bottleneck perceptron, and uses adversarial contrastive learning for training.
result DefNet significantly improves GNN robustness under various adversarial attacks.

ADMP-GNN dynamically adjusts message-passing layers for better graph learning performance.

problem Fixed message-passing steps in GNNs do not account for nodes' varying computational needs.
method Proposes ADMP-GNN, which dynamically adjusts the number of message-passing layers for each node.
result Improves performance on node classification tasks compared to baseline GNN models.

CI-GNN uses GNNs to diagnose psychiatric disorders by identifying causally relevant brain regions.

problem Leveraging GNNs for psychiatric diagnosis requires interpretable models to understand decision-making.
method CI-GNN integrates Granger causality into GNNs to identify causally relevant subgraphs.
result CI-GNN provides more reliable and concise explanations of psychiatric diagnoses.

This Ph.D. thesis is devoted to the constructions of Lagrangian formulation on Finsler and Kawaguchi manifolds. While Finsler geometry is a natural extension of Riemannian geometry, Kawaguchi geometry is the extension of Finsler geometry to higher order derivatives and to k-dimensional parameter space. The latter exten…

2013-10-16abs ↗pdf ↗

We prove that any asymptotically locally Euclidean scalar-flat Kähler 4-orbifold whose isometry group contains a 2-torus is isometric, up to an orbifold covering, to a quaternionic-complex quotient of a kk-dimensional quaternionic vector space by a (k1)(k-1)-torus. In order to do so, we first prove that any compact anti…

2009-02-10abs ↗pdf ↗

Paper tackles fairness issues in GNNs by proposing ELEGANT for certification.

problem Fairness issues in GNN predictions due to graph data perturbations.
method Proposes ELEGANT framework for certifying fairness of any GNN without assumptions or re-training.
result The fairness of any GNN backbone is impossible to be corrupted under certain perturbation budgets.

We show that for closed orientable manifolds the kk-dimensional stable systole admits a metric-independent volume bound if and only if there are cohomology classes of degree kk that generate cohomology in top-degree. Moreover, it turns out that in the nonorientable case such a bound does not exist for stable systoles…

2007-08-20abs ↗pdf ↗

GraphNorm accelerates GNN training by adapting InstanceNorm, improving convergence and generalization.

problem Improving convergence and generalization of Graph Neural Networks (GNNs).
method Adapting InstanceNorm to GNNs, proposing GraphNorm with a learnable shift.
result GNNs with GraphNorm converge faster and achieve better performance on benchmarks.

Boost GNNs for node classification by incorporating label dependencies.

problem Current GNNs lack expressiveness and fail to capture label dependencies.
method Proposes a collective learning framework combining collective classification and self-supervised learning.
result Consistent, significant improvement in node classification accuracy across various GNNs.

AGNN automates GNN architecture search, achieving best performance.

problem Finding optimal GNN architectures is laborious and requires human expertise.
method AGNN uses reinforcement learning to search for optimal GNN architectures within a predefined space, with a novel parameter sharing strategy.
result AGNN identifies optimal GNN architectures achieving best performance.

SGQuant reduces GNN memory usage without significant accuracy loss.

problem High memory consumption in GNNs limits their applicability on memory-constrained devices.
method Proposes a specialized GNN quantization scheme (SGQuant) with a quantization algorithm, fine-tuning scheme, and multi-granularity strategy.
result SGQuant reduces GNN memory footprint from 4.25x to 31.9x with minimal accuracy loss.

New research limits what GNNs can compute and generalizes their performance.

problem Limits of GNNs in computing graph properties and generalization bounds.
method Novel graph-theoretic formalism and data-dependent generalization bounds.
result Proves GNNs can't compute certain graph properties and provides tighter generalization bounds.

PA-GNN enhances GNN robustness against poisoning attacks using clean graph knowledge.

problem Improving robustness of GNNs against poisoning attacks.
method PA-GNN uses a penalized aggregation mechanism and meta-optimization to transfer robustness from clean graphs.
result PA-GNN significantly improves GNN robustness against poisoning attacks on real-world graphs.

Alt-GNNs improve travel mode choice modeling by integrating graph neural networks with GEV models.

problem Capturing alternative dependence in discrete choice models with predefined, symmetric, and uniform dependence.
method Introducing Alternative Graph Neural Networks (Alt-GNNs) that embed alternative dependence within a unified framework.
result Alt-GNNs significantly improve predictive performance over benchmark models in travel mode choice datasets.