skills.rest
hiyenwong
hiyenwongCOMMUNITY
Explore all skills by this author →
𝕏@khonsu_51GitHub

specificity-aware-federated-graph-learning

Train federated graph neural networks on multi-site fMRI data with shared and personalized branches.

2|Updated Feb 12, 2026
Search Tags:#privacy#graph-neural-networks#multi-site#fMRI#federated-learning#brain-imaging
One-click install
npx skills add https://github.com/hiyenwong/ai_collection --skill specificity-aware-federated-graph-learning
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: specificity-aware-federated-graph-learning
Source: https://github.com/hiyenwong/ai_collection/tree/main/collection/skills/specificity-aware-federated-graph-learning
Command: npx skills add https://github.com/hiyenwong/ai_collection --skill specificity-aware-federated-graph-learning

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Privacy-preserving federated learning for multi-site fMRI data to balance shared knowledge with site-specific patterns.

Core Features & Use Cases

  • Shared knowledge extraction via a common graph model
  • Personalization to preserve site-specific demographics and connectivity
  • Use cases include multi-center brain disease identification and rs-fMRI analysis, enabling privacy-preserving collaboration

Quick Start

Provide multi-site fMRI data and start the federated training run with shared and personalized branches.

Frequently Asked Questions about specificity-aware-federated-graph-learning

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run federated learning on multi-site fMRI data without sharing raw brain imaging scans?▼

Federated learning on multi-site fMRI data keeps raw scans local by running client-side training loops and aggregating only model updates server-side using FedAvg. This balances shared knowledge extraction with site-specific personalization across multiple centers.

What is specificity-aware federated graph learning for brain disease identification?▼

Specificity-aware federated graph learning uses a graph neural network with a shared branch and a personalized branch. This mechanism extracts common functional connectivity patterns while preserving site-specific demographics for multi-center brain disease identification.

Can I use graph neural networks for rs-fMRI analysis across centers with different demographics?▼

Yes, graph neural networks can analyze rs-fMRI across centers with varying demographics by using a personalized branch to accommodate site-specific functional connectivity. This approach balances shared knowledge with local patterns during federated training.

What's the best way to preserve site-specific patterns in multi-site fMRI federated learning?▼

The best way to preserve site-specific patterns in multi-site fMRI federated learning is using a dual-branch graph neural network. A shared branch extracts common features while a personalized branch retains local demographics and connectivity, aggregated via server-side FedAvg.

Does federated graph learning require raw fMRI data to be transferred to a central server?▼

No, federated graph learning does not require transferring raw fMRI data to a central server. Raw data remains local, and only model updates from the shared and personalized branches are aggregated server-side using FedAvg for privacy-preserving collaboration.

Why does my multi-site fMRI model fail to generalize across different brain imaging centers?▼

Multi-site fMRI models often fail to generalize due to demographic variation and differing functional connectivity across centers. Using a specificity-aware federated graph learning approach with personalized branches addresses this by accommodating site-specific patterns alongside shared knowledge.

Related Skills

Explore More Education & Research Skills→
hiyenwonghiyenwong
2

atlas-free-brain-network-transformer

Generate subject-level embeddings from rs-fMRI data using atlas-free parcellation.

Community
Advanced
hiyenwonghiyenwong
2

dgcl-brain-network-construction

Automate end-to-end brain network construction from fMRI data using DGCL.

Community
Advanced
hiyenwonghiyenwong
2

federated-brain-trajectory-gnn

Predict infant brain connectivity evolution with federated multi-trajectory GNNs.

Community
Advanced
robotlearning123robotlearning123

torch-geometric

Develop and train graph neural networks with PyTorch.

Community
Advanced