Add MDP (Matrix Decomposition + Differential Privacy) defense#35
Open
Jeneidi wants to merge 1 commit intoLabRAI:mainfrom
Open
Add MDP (Matrix Decomposition + Differential Privacy) defense#35Jeneidi wants to merge 1 commit intoLabRAI:mainfrom
Jeneidi wants to merge 1 commit intoLabRAI:mainfrom
Conversation
This adds a new privacy-preserving defense mechanism for GNNs that: - Splits adjacency matrix into shares via eigendecomposition - Applies Laplace noise to features for differential privacy - Uses federated training with parameter averaging across calculators New files: - pygip/utils/mdp/ - Core MDP utilities (abar, eigenvalue_split, dp_features, splits) - pygip/models/nn/mdp_gcn.py - ManualGCN model for dense adjacency matrices - pygip/models/defense/MDP.py - Main defense class extending BaseDefense - examples/defense/MDP.py - Usage example
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a new privacy-preserving defense mechanism for GNNs:
Files Added
Usage
from pygip.datasets import Cora
from pygip.models.defense import MDP
dataset = Cora(api_type='pyg')
defense = MDP(dataset, nc=4, es=2, epsilon=30.0)
res, res_comp = defense.defend()