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.
GWNN uses graph wavelets for efficient graph CNNs.
New method selects diffusion scales for graph wavelets.
We propose a new framework for manifold denoising based on processing in the graph Fourier frequency domain, derived from the spectral decomposition of the discrete graph Laplacian. Our approach uses the Spectral Graph Wavelet transform in order to per- form non-iterative denoising directly in the graph frequency domai…
Graph scattering transforms are stable to metric perturbations of network topology.
Proposes a method to adapt labels on graphs with few labeled nodes.
Proposes a new dictionary learning method for high-dimensional graph signals.
New spectral triples for higher-rank graphs linked to wavelet decompositions.
A new GNN module learns geometric scattering features for better graph classification and feature exploration.
Graph classification improved using spectral features and wavelet filters.
Harmonic analysis on directed graphs for signal modeling and semi-supervised learning.
SpGAT learns graph representations using spectral attention for efficiency.