Papers › Efficient nonparametric statistical inference on population feature importance using...

Efficient nonparametric statistical inference on population feature importance using Shapley values

16 Jun 2020ICML 2020 1arXiv:2006.09481archive 2025-07-28

Brian D. Williamson, Jean Feng

The true population-level importance of a variable in a prediction task provides useful knowledge about the underlying data-generating mechanism and can help in deciding which measurements to collect in subsequent experiments. Valid statistical inference on this importance is a key component in understanding the population of interest. We present a computationally efficient procedure for estimating and obtaining valid statistical inference on the Shapley Population Variable Importance Measure (SPVIM). Although the computational complexity of the true SPVIM scales exponentially with the number of variables, we propose an estimator based on randomly sampling only Θ(n) feature subsets given n observations. We prove that our estimator converges at an asymptotically optimal rate. Moreover, by deriving the asymptotic distribution of our estimator, we construct valid confidence intervals and hypothesis tests. Our procedure has good finite-sample performance in simulations, and for an in-hospital mortality prediction task produces similar variable importance estimates when different machine learning algorithms are applied.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

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.

Syntology holds a reference implementation of Shapley Population Variable Importance Measure estimator, generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='Shapley Population Variable Importance Measure estimator') (how to connect); the free check have('2006.09481') needs no token.

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

Code

Syntology 25 code samples harvested from 2 repositories linked to this paper: 15 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 13 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; 12 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 25 samples from 2 repositories, 15 ran (2 with no instrument failure; 13 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.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

bdwilliamson/spvim_supplementary mentioned on GitHubSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
bdwilliamson/vimp mentioned on GitHubNOASSERTIONSyntology: not harvested report
bdwilliamson/vimpy mentioned on GitHubSyntology: 24 samples, 14 ran (12 instrument-failed) · 10 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

25 samples harvested: 15 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 13 where Syntology's instrument failed) · 10 unverified. “Where Syntology's instrument failed” counts failures 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.

2ran · honoured contract
1ran · our draft was wrong
12ran; the contract check could not run
10unverified

Licence: 0 of the 25 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 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

bdwilliamson/spvim_supplementary community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
shapley_influence_function code/data_analysis/get_influence_functions.py ran; the contract check could not run (RAISES) MIT (permissive) · 610ae143866b4724 · report fixture: signature Z:ndarray/2/float, z_counts:ndarray/1/int, W:ndarray/2/float, v:ndarray/1/float, psi:ndarray/1/float, G:ndarray/2/float, c_n:ndarray/1/float, ics:ndarray/1/float, measure:str/0/str · returned dict · level recorded 2026-09-22
bdwilliamson/vimpy community (archive-listed) 24 samples, 14 ran (12 instrument-failed) · 10 unverified
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
vimpy/spvim.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 5efa7eb653ea329a · report fixture: signature y:ndarray/1/int, preds:ndarray/2/int · returned list · level recorded 2026-09-22
auc in 21 papers vimpy/spvim.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · b3b4f79c3ab9cdf9 · report fixture: signature y:ndarray/1/int, preds:ndarray/1/float · returned float · level recorded 2026-09-22
auc_ic vimpy/spvim.py ran; the contract check could not run (RAISES) MIT (permissive) · 514d20b055ecbe1b · report fixture: signature y:ndarray/1/int, preds:ndarray/2/float · returned ndarray · level recorded 2026-09-24
choose in 11 papers vimpy/spvim.py ran · honoured contract fingerprinted MIT (permissive) · 96c2d58d78728d79 · report fixture: signature n:int/0/int, k:int/0/int · returned int · level recorded 2026-09-24
cross_entropy in 92 papers
census: 6 implementations from 8 papers, 5 distinct outputs across 2 buckets, one shared input per bucket
vimpy/spvim.py ran; the contract check could not run (RAISES) MIT (permissive) · 64ed85c06ce35e76 · report fixture: signature y:ndarray/1/int, preds:ndarray/2/float · returned list · level recorded 2026-09-24
deviance vimpy/spvim.py ran; the contract check could not run (RAISES) MIT (permissive) · f22751c60cbae69a · report fixture: signature y:ndarray/2/float, preds:ndarray/2/float · returned list · level recorded 2026-09-24
make_folds vimpy/spvim.py ran; the contract check could not run (RAISES) MIT (permissive) · a6fd8ce09cf07e2f · report fixture: signature y:ndarray/1/int, V:int/0/int, stratified:bool/0/bool · returned ndarray · level recorded 2026-09-24
one_auc_ic vimpy/spvim.py ran; the contract check could not run (RAISES) MIT (permissive) · 315e68efb0762ec6 · report fixture: signature y:ndarray/1/int, preds:ndarray/1/float · returned ndarray · level recorded 2026-09-24
one_r2_ic vimpy/spvim.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 38bb504d66f788f8 · report fixture: signature y:ndarray/1/float, preds:ndarray/1/float · returned ndarray · level recorded 2026-09-24
r_squared in 5 papers vimpy/spvim.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 534144f19cbd0ac6 · report fixture: signature y:ndarray/1/float, preds:ndarray/2/float · returned list · level recorded 2026-09-22
r_squared_ic vimpy/spvim.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 85ce6da752e9ae14 · report fixture: signature y:ndarray/1/float, preds:ndarray/2/float · returned ndarray · level recorded 2026-09-22
shapley_hyp_test vimpy/spvim.py ran · our draft was wrong fingerprinted MIT (permissive) · 8e32ea0b9aa76ef2 · report fixture: signature vs_one_1:ndarray/1/float, v_none_0:float/0/float, sigmas_one:ndarray/1/float, sigma_none:float/0/float, delta:int/0/int, level:float/0/float, p:int/0/int · returned tuple · level recorded 2026-09-22
shapley_influence_function vimpy/spvim.py ran; the contract check could not run (RAISES) MIT (permissive) · 8468321fa0661658 · report fixture: signature Z:ndarray/2/float, z_counts:ndarray/1/int, W:ndarray/2/float, v:ndarray/1/float, psi:ndarray/1/float, G:ndarray/2/float, c_n:ndarray/1/float, ics:ndarray/1/float, measure:str/0/str · returned dict · level recorded 2026-09-24
shapley_se vimpy/spvim.py ran · honoured contract MIT (permissive) · 98c482a48f51a3d5 · report fixture: signature shapley_ics:dict/0/?, idx:int/0/int, gamma:float/0/float, na_rm:bool/0/bool · returned float64 · level recorded 2026-09-22
accuracy_ic vimpy/spvim.py unverified MIT (permissive) · af13607054544c80 · report
compute_ic vimpy/spvim.py unverified MIT (permissive) · 559611a6ba26946d · report
cross_entropy_ic vimpy/spvim.py unverified MIT (permissive) · f40e6267444d9487 · report
cv_predictiveness vimpy/spvim.py unverified MIT (permissive) · 1c5b419e1c08be46 · report
deviance_ic vimpy/spvim.py unverified MIT (permissive) · b03f39be86306d13 · report
get_measure_function vimpy/spvim.py unverified MIT (permissive) · 7834085f6f2c900c · report
one_accuracy_ic vimpy/spvim.py unverified MIT (permissive) · 6e65685b71563fac · report
one_cross_entropy_ic vimpy/spvim.py unverified MIT (permissive) · 2d1e9e7840f75757 · report
one_deviance_ic vimpy/spvim.py unverified MIT (permissive) · 434a920696b14b3d · report
spvim vimpy/spvim.py unverified MIT (permissive) · f65b08b4141edc84 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another. Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

3 assemblies on this paper: 2 harvested, 1 generated, listed in 2 groups (siblings lifted in one batch form one group).

generated Assembly for Shapley Population Variable Importance Measure estimator (graph method: Shapley Population Variable Importance Measure estimator), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 3; 3 property tests passed on both).

label matches by construction: the code was generated for this method (not judged by a model)

  1. solve_spvim_from_predictiveness entry · level 3 · ran; the contract check could not run · 1cdb9bc1efc1abd8cb91d44e11efd3a8ca1321c44c067086c4c917239146c122

2 sibling assemblies for Shapley Population Variable Importance Measure estimator (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Shapley Population Variable Importance Measure estimator (graph method: Shapley Population Variable Importance Measure estimator). 1 brick; individually on a fixture: 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. The assembly has not been run end-to-end.

one of 2 siblings

  1. shapley_influence_function entry · level 1 · ran; the contract check could not run (RAISES) · 610ae143866b4724

harvested Assembly for Shapley Population Variable Importance Measure estimator (graph method: Shapley Population Variable Importance Measure estimator). 24 bricks; individually on a fixture: 14 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 12 where Syntology's instrument failed) · 10 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. spvim entry · level 0 · unverified · f65b08b4141edc84
  2. choose helper · level 1 · ran · honoured contract · 96c2d58d78728d79
  3. compute_ic helper · level 0 · unverified · 559611a6ba26946d
  4. cv_predictiveness model · level 0 · unverified · 1c5b419e1c08be46
  5. get_measure_function helper · level 0 · unverified · 7834085f6f2c900c
  6. shapley_hyp_test helper · level 1 · ran · our draft was wrong · 8e32ea0b9aa76ef2
  7. shapley_influence_function helper · level 1 · ran; the contract check could not run (RAISES) · 8468321fa0661658
  8. shapley_se helper · level 1 · ran · honoured contract · 98c482a48f51a3d5
  9. accuracy_ic metric · level 0 · unverified · af13607054544c80
  10. auc_ic metric · level 1 · ran; the contract check could not run (RAISES) · 514d20b055ecbe1b
  11. cross_entropy_ic metric · level 0 · unverified · f40e6267444d9487
  12. deviance_ic metric · level 0 · unverified · b03f39be86306d13
  13. r_squared_ic metric · level 1 · ran; the contract check could not run (RAISES) · 85ce6da752e9ae14
  14. make_folds helper · level 1 · ran; the contract check could not run (RAISES) · a6fd8ce09cf07e2f
  15. accuracy helper · level 1 · ran; the contract check could not run (RAISES) · 5efa7eb653ea329a
  16. auc helper · level 1 · ran; the contract check could not run (RAISES) · b3b4f79c3ab9cdf9
  17. deviance helper · level 1 · ran; the contract check could not run (RAISES) · f22751c60cbae69a
  18. r_squared helper · level 1 · ran; the contract check could not run (RAISES) · 534144f19cbd0ac6
  19. one_accuracy_ic helper · level 0 · unverified · 6e65685b71563fac
  20. one_auc_ic helper · level 1 · ran; the contract check could not run (RAISES) · 315e68efb0762ec6
  21. one_cross_entropy_ic helper · level 0 · unverified · 2d1e9e7840f75757
  22. one_deviance_ic helper · level 0 · unverified · 434a920696b14b3d
  23. one_r2_ic helper · level 1 · ran; the contract check could not run (RAISES) · 38bb504d66f788f8
  24. cross_entropy helper · level 1 · ran; the contract check could not run (RAISES) · 64ed85c06ce35e76

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2006.09481") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Feature ImportanceMortality Prediction

1 archive task tag without a task page not shown.

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