Papers › Provable Adversarial Robustness for Group Equivariant Tasks: Graphs, Point Clouds,...

Provable Adversarial Robustness for Group Equivariant Tasks: Graphs, Point Clouds, Molecules, and More

5 Dec 2023NeurIPS 2023 11arXiv:2312.02708archive 2025-07-28

Jan Schuchardt, Yan Scholten, Stephan Günnemann

A machine learning model is traditionally considered robust if its prediction remains (almost) constant under input perturbations with small norm. However, real-world tasks like molecular property prediction or point cloud segmentation have inherent equivariances, such as rotation or permutation equivariance. In such tasks, even perturbations with large norm do not necessarily change an input's semantic content. Furthermore, there are perturbations for which a model's prediction explicitly needs to change. For the first time, we propose a sound notion of adversarial robustness that accounts for task equivariance. We then demonstrate that provable robustness can be achieved by (1) choosing a model that matches the task's equivariances (2) certifying traditional adversarial robustness. Certification methods are, however, unavailable for many models, such as those with continuous equivariances. We close this gap by developing the framework of equivariance-preserving randomized smoothing, which enables architecture-agnostic certification. We additionally derive the first architecture-specific graph edit distance certificates, i.e. sound robustness guarantees for isomorphism equivariant tasks like node classification. Overall, a sound notion of robustness is an important prerequisite for future work at the intersection of robust and geometric machine learning.

PaperPDFConference PDFCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2312.02708")

Code

Syntology Ran 22 of 31 code samples harvested from 5 repositories linked to this paper; 9 have no recorded run. Of those that ran: 2 ran · our draft was wrong; 1 ran · fixture could not drive it; 19 ran with no contract checked.

By repository: found in paper text by Syntology: 29 samples from 5 repositories, 20 ran; 2 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

danielzuegner/robust-gcn found in paper text by Syntology report
RobustGraph/RoboGraph found in paper text by SyntologyMIT report
abojchevski/graph_cert found in paper text by Syntology report
abojchevski/sparse_smoothing found in paper text by Syntology report
jan-schuchardt/invariance-smoothing found in paper text by SyntologyMIT report
wollschl/uncertainty_for_molecules found in paper text by Syntology report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

31 samples harvested; 22 ran; 0 honoured the contract we drafted; 9 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · our draft was wrong
1ran · fixture could not drive it
19ran
9unverified

Licence: 4 of the 31 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

RobustGCNLayer danielzuegner/robust-gcn/robust_gcn/robust_gcn.py found in paper text by Syntology ran MIT (permissive) · a268573b18831a29 · report
SPG RobustGraph/RoboGraph/robograph/attack/SPG.py found in paper text by Syntology ran MIT (permissive) · 9073926e56ac1c34 · report
add_gaussian_noise jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/training.py found in paper text by Syntology ran MIT (permissive) · fee01cc7d1d5c624 · report
calculate_budget RobustGraph/RoboGraph/robograph/attack/genetic_attack.py found in paper text by Syntology ran MIT (permissive) · 637b58808cd09ffa · report
check_local RobustGraph/RoboGraph/robograph/attack/genetic_attack.py found in paper text by Syntology ran MIT (permissive) · 313a20ad80cd5b01 · report
closed_form_B RobustGraph/RoboGraph/robograph/attack/admm.py found in paper text by Syntology ran MIT (permissive) · a01aa6ebe170c0dc · report
compute_rho abojchevski/sparse_smoothing/sparse_smoothing/cert.py found in paper text by Syntology ran · fixture could not drive it fingerprinted no licence file found · pointer only · 13a44cb31287e83c · report
dict_to_dot jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/utils.py found in paper text by Syntology ran MIT (permissive) · 2d5870b6c4fe5b59 · report
eval RobustGraph/RoboGraph/robograph/model/gnn.py found in paper text by Syntology ran MIT (permissive) · 4bae3496a287a478 · report
forward_cert_cohen jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/certificates/blackbox.py found in paper text by Syntology ran fingerprinted MIT (permissive) · de80f2bd3a451839 · report
inverse_cert_cohen jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/certificates/blackbox.py found in paper text by Syntology ran fingerprinted MIT (permissive) · e2bcee4d8b42f4b1 · report
isLegal RobustGraph/RoboGraph/robograph/attack/SPG.py found in paper text by Syntology ran MIT (permissive) · 02d6a2dca6a6dc32 · report
load_graph RobustGraph/RoboGraph/robograph/utils.py found in paper text by Syntology ran MIT (permissive) · b6d7cd2f3295bedf · report
local_solver RobustGraph/RoboGraph/robograph/attack/dp_old_version.py found in paper text by Syntology ran MIT (permissive) · dc92a95e2bacf927 · report
normalize_adj RobustGraph/RoboGraph/robograph/model/gcn_conv.py found in paper text by Syntology ran fingerprinted MIT (permissive) · c55badf69fe079ef · report
polyinterp2 RobustGraph/RoboGraph/robograph/attack/SPG.py found in paper text by Syntology ran fingerprinted MIT (permissive) · 4d997c194830837b · report
process_data RobustGraph/RoboGraph/robograph/utils.py found in paper text by Syntology ran MIT (permissive) · 802848e9f51394d1 · report
randomly_scale jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/training.py found in paper text by Syntology ran MIT (permissive) · be9f96362f6c27b7 · report
sample_single_axis_likelihood_ratio_logs jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/certificates/rotation_elemental.py found in paper text by Syntology ran MIT (permissive) · 1daf26cd6d81ba89 · report
train RobustGraph/RoboGraph/robograph/model/gnn.py found in paper text by Syntology ran MIT (permissive) · 3df5b2fc12d5a4ec · report
RobustGCNModel danielzuegner/robust-gcn/robust_gcn/robust_gcn.py found in paper text by Syntology unverified MIT (permissive) · a3af207d6d21af5d · report
attack_global RobustGraph/RoboGraph/robograph/utils.py found in paper text by Syntology unverified MIT (permissive) · 98474b68028dbdbe · report
dp_solver RobustGraph/RoboGraph/robograph/attack/dp_old_version.py found in paper text by Syntology unverified MIT (permissive) · 714d5d63e4a9394b · report
forward_elemental_rotation_cert jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/certificates/rotation_elemental.py found in paper text by Syntology unverified MIT (permissive) · f60179fa8fb21d2b · report
inverse_elemental_rotation_cert jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/certificates/rotation_elemental.py found in paper text by Syntology unverified MIT (permissive) · 015c7f55ac954727 · report
local_solver_linear_term RobustGraph/RoboGraph/robograph/attack/dp_old_version.py found in paper text by Syntology unverified MIT (permissive) · 5ad2414d2eb09c07 · report
sample_votes_datapoint jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/smoothing.py found in paper text by Syntology unverified MIT (permissive) · 97b718aa6a264cde · report
sample_votes_dataset jan-schuchardt/invariance-smoothing/pointcloud_invariance_smoothing/smoothing.py found in paper text by Syntology unverified MIT (permissive) · c42e1cd047ffdb2c · report
wrap_prediction_model wollschl/uncertainty_for_molecules/ue4mol/models/utils.py found in paper text by Syntology unverified no licence file found · pointer only · 7510b92771842954 · report
preprocess_adj identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · a3cccfaa8bd4c86c · report
sparse_tensor identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · c25e88689b0251b5 · report

Tasks

Adversarial RobustnessMolecular Property PredictionNode ClassificationPoint Cloud SegmentationPredictionProperty Prediction

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections