arXiv Statistics ML
Outlier Detection for Multi-Network Data
Read the original on arXiv Statistics ML →The Flow has not summarised this story yet — read it at arXiv Statistics ML.
Related stories
arXiv Statistics ML
Sep 15
Assumption-Lean Inference for Spectral Differential Network Analysis of High-Dimensional Time Series
arXiv:2609.13609v1 Announce Type: cross Abstract: Network analysis for multivariate time series is popular in many fields, from neuroscience to seismology. The inverse spectral density is a common ch...
arXiv AI
Sep 17
A Non-Linear Neuron Based Detection of Isolated Pixels in Binary and Grayscale Images using Contrast Sensitive Receptive Fields
The paper presents a new method for detecting isolated pixels in both binary and grayscale images by extending a neuron-based anomaly detection model to use contrast-sensitive receptive fields with excitatory and inhibitory regions. It addresses limitations of existing techniques such as template matching and second‑order derivative methods, which are either infeasible for grayscale images or overly sensitive to noise and require user‑defined thresholds. The proposed approach eliminates the need for user‑specified parameters, offering a robust, efficient solution applicable across various image processing domains.
arXiv Machine Learning
Jun 30
Connectivity Estimation using Stochastic Graph Heat Modelling
arXiv:2606. 29098v1 Announce Type: cross Abstract: A growing number of techniques leverage the spatial structures that underlie many real-world datasets.
arXiv Machine Learning
Jul 14
MVMGNN;Multi-View Masked Graph Neural Network for Alzheimer's Disease Diagnosis using Structural MRI
arXiv:2607. 09788v1 Announce Type: cross Abstract: Alzheimer's disease (AD) is a common neurodegenerative disorder, and early diagnosis is of great significance for delaying disease progression and enabling timely intervention.
arXiv Machine Learning
Aug 31
Multiscale Community-Based Fingerprinting of Signed Functional Networks
The paper introduces a multiscale community-based fingerprinting framework for signed functional brain networks, using a signed multilayer community detection approach that captures both correlated and anti-correlated activity. Graph-theoretic metrics derived from the joint community structures yield low-dimensional, interpretable fingerprints that reliably identify individuals across multiple sessions and tasks. Evaluation on 810 healthy controls from the Human Connectome Project demonstrates that these community-based fingerprints outperform traditional edge-level methods in stability and interpretability.
arXiv Machine Learning
Aug 26
Multi-Source Complex Network Reconstruction via Wasserstein Distributionally Robust Optimization and Algorithm Unrolling
The paper introduces MS‑WDRO, a multi‑source Wasserstein distributionally robust optimization framework for reconstructing complex network topologies from scarce target‑domain data and abundant heterogeneous source data. It fuses sources via a weighted Wasserstein barycenter, builds an ambiguity set around it, and solves a regularized Laplacian estimator using a provably convergent ADMM scheme. The authors provide finite‑sample guarantees, demonstrate that naive aggregation is suboptimal, and show through experiments on synthetic data and the ABIDE I neuroimaging dataset that MS‑WDRO outperforms seven baselines in graph recovery, sample efficiency, and diagnostic utility, especially when target samples are limited.