Documentation: API reference, developer guide, expanded theory and tutorials#297
Merged
Kiguli merged 4 commits intoverivital:masterfrom Apr 2, 2026
Merged
Documentation: API reference, developer guide, expanded theory and tutorials#297Kiguli merged 4 commits intoverivital:masterfrom
Kiguli merged 4 commits intoverivital:masterfrom
Conversation
Add comprehensive theory page for Graph Neural Network verification covering GraphStar sets, GCN/GINE reachability, subgraph verification, soundness guarantees, and experimental results summary. Based on the GNNV paper (under review). Also fix RST indentation warning in acasxu.rst.
Theory pages: - GNN reachability: GraphStar sets, GCN/GINE reachability, subgraph verification, soundness guarantees (based on GNNV paper, under review) - Weight perturbation (ModelStar): perturbation maps, exact/over-approximate reachability, quantization connection, INN relationship (based on Zubair et al. IEEE CAI 2025 + JAIR accepted) - Probabilistic verification: substantially expanded with double-step PAC guarantee, nonconformity scores, surrogate-based approach, deflative PCA, pixel-level classification, randomized smoothing comparison, experimental results (based on Hashemi et al. NeurIPS 2025) Tutorial: - Weight perturbation example with quantization interpretation table
Theory pages now contain only theoretical foundations: - fairness.md: removed results table and key observations - probabilistic.md: removed experimental summary table and comparison numbers - gnn-reachability.md: removed experimental results summary Example/tutorial pages now contain all experimental content: - NEW fairness.rst: complete FairNNV tutorial with results table, debiasing paradox finding, and VF score walkthrough - semantic-segmentation.rst: added probabilistic verification results table (M2NIST, Lung Seg, OCTA-500, CamVid, Cityscapes) with comparison numbers - gnn.rst: added experimental results summary with CORA comparison
…veloper guide New sections: - API Reference: NN class, GNN class, set types, 48 layer types, NNCS classes, utilities, and reachOptions struct reference - Developer Guide: architecture overview, custom verification workflows, adding new layers/sets/methods, testing & CI/CD Navigation improvements: - Developer Guide appears before API Reference in sidebar - Testing section merged into Developer Guide (developer-only content) - Courses page moved to bottom of main menu as standalone item - Copyright updated to 2026 API accuracy fixes (verified against source code): - verify_vnnlib returns [result, X] not just result - verify_specification: 0=failed, 1=satisfied, 2=unknown - matlab2nnv also accepts LayerGraph - WPutils is a static class with weightPerturb property - Added verify_safety and verify_segmentation to NN methods
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
Significant expansion of the GitHub Pages documentation site at https://verivital.github.io/nnv/ with new sections for API reference, developer extensibility guide, and expanded theory/tutorial content based on NNV3 feature papers.
New sections
Expanded theory pages (based on published papers)
New tutorials
Structural improvements
Test plan
sphinx-buildpasses with zero warnings