Papers › Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure

Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure

13 Jun 2022arXiv:2206.06219archive 2025-07-28

Paul Novello, Thomas Fel, David Vigouroux

This paper presents a new efficient black-box attribution method based on Hilbert-Schmidt Independence Criterion (HSIC), a dependence measure based on Reproducing Kernel Hilbert Spaces (RKHS). HSIC measures the dependence between regions of an input image and the output of a model based on kernel embeddings of distributions. It thus provides explanations enriched by RKHS representation capabilities. HSIC can be estimated very efficiently, significantly reducing the computational cost compared to other black-box attribution methods. Our experiments show that HSIC is up to 8 times faster than the previous best black-box attribution methods while being as faithful. Indeed, we improve or match the state-of-the-art of both black-box and white-box attribution methods for several fidelity metrics on Imagenet with various recent model architectures. Importantly, we show that these advances can be transposed to efficiently and faithfully explain object detection models such as YOLOv4. Finally, we extend the traditional attribution methods by proposing a new kernel enabling an ANOVA-like orthogonal decomposition of importance scores based on HSIC, allowing us to evaluate not only the importance of each image patch but also the importance of their pairwise interactions. Our implementation is available at https://github.com/paulnovello/HSIC-Attribution-Method.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 1 code sample harvested from 1 repository linked to this paper: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code. Every run on this page is a failure of Syntology's instrument, not of the code.

By repository: official repository: 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No date is recorded for the sample that ran: its level was written before Syntology's graph kept the date.

paulnovello/hsic-attribution-method officialmentioned in papertfSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified 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

1 sample harvested: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Every run on this page is a failure of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong

Licence: 1 of the 1 sample is 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 paulnovello/hsic-attribution-method. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

paulnovello/hsic-attribution-method official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
get_alpha_cmap utils.py ran · our draft was wrong no licence file found · pointer only · 8bce65a80b5ee368 · report fixture: signature cmap:str/0/str · returned ListedColormap · level recorded, date not recorded

Tasks

Error UnderstandingImage AttributionInterpretability Techniques for Deep LearningObject Detectionobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Error Understanding CUB-200-2011 HSIC-Attribution Average highest confidence (EfficientNetV2-M) 0.2679 #4 of 4 Archive leaderboard report
Error Understanding CUB-200-2011 HSIC-Attribution Average highest confidence (MobileNetV2) 0.2914 #4 of 4 Archive leaderboard report
Error Understanding CUB-200-2011 HSIC-Attribution Average highest confidence (ResNet-101) 0.2493 #4 of 4 Archive leaderboard report
Error Understanding CUB-200-2011 HSIC-Attribution Insertion AUC score (EfficientNetV2-M) 0.1611 #4 of 4 Archive leaderboard report
Error Understanding CUB-200-2011 HSIC-Attribution Insertion AUC score (MobileNetV2) 0.1635 #4 of 4 Archive leaderboard report
Error Understanding CUB-200-2011 HSIC-Attribution Insertion AUC score (ResNet-101) 0.1446 #4 of 4 Archive leaderboard report
Error Understanding CUB-200-2011 (ResNet-101) HSIC-Attribution Average highest confidence 0.2493 #3 of 3 Archive leaderboard report
Error Understanding CUB-200-2011 (ResNet-101) HSIC-Attribution Insertion AUC score 0.1446 #3 of 3 Archive leaderboard report
Image Attribution CUB-200-2011 HSIC-Attribution Deletion AUC score (ResNet-101) 0.0647 #4 of 8 Archive leaderboard report
Image Attribution CUB-200-2011 HSIC-Attribution Insertion AUC score (ResNet-101) 0.6843 #4 of 8 Archive leaderboard report
Image Attribution CelebA HSIC-Attribution Deletion AUC score (ArcFace ResNet-101) 0.1151 #3 of 8 Archive leaderboard report
Image Attribution CelebA HSIC-Attribution Insertion AUC score (ArcFace ResNet-101) 0.5692 #3 of 8 Archive leaderboard report
Image Attribution VGGFace2 HSIC-Attribution Deletion AUC score (ArcFace ResNet-101) 0.1317 #2 of 8 Archive leaderboard report
Image Attribution VGGFace2 HSIC-Attribution Insertion AUC score (ArcFace ResNet-101) 0.6694 #2 of 8 Archive leaderboard report
Interpretability Techniques for Deep Learning CelebA HSIC-Attribution Insertion AUC score 0.5692 #2 of 7 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 ConvolutionAverage PoolingBatch NormalizationBottom-up Path AugmentationCSPDarknet53ConvolutionCosine AnnealingCutMixDropBlockFPNGlobal Average PoolingGrid SensitiveLabel SmoothingLogistic RegressionMax PoolingPAFPNReLUResidual ConnectionSigmoid ActivationSoftmaxSpatial Pyramid PoolingTanh ActivationYOLOv3YOLOv4k-Means Clustering

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