Papers › Principal Neighbourhood Aggregation for Graph Nets

Principal Neighbourhood Aggregation for Graph Nets

12 Apr 2020NeurIPS 2020 12arXiv:2004.05718archive 2025-07-28

Gabriele Corso, Luca Cavalleri, Dominique Beaini, Pietro Liò, Petar Veličković

Graph Neural Networks (GNNs) have been shown to be effective models for different predictive tasks on graph-structured data. Recent work on their expressive power has focused on isomorphism tasks and countable feature spaces. We extend this theoretical framework to include continuous features - which occur regularly in real-world input domains and within the hidden layers of GNNs - and we demonstrate the requirement for multiple aggregation functions in this context. Accordingly, we propose Principal Neighbourhood Aggregation (PNA), a novel architecture combining multiple aggregators with degree-scalers (which generalize the sum aggregator). Finally, we compare the capacity of different models to capture and exploit the graph structure via a novel benchmark containing multiple tasks taken from classical graph theory, alongside existing benchmarks from real-world domains, all of which demonstrate the strength of our model. With this work, we hope to steer some of the GNN research towards new aggregation methods which we believe are essential in the search for powerful and robust models.

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.

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

Code

No sample from the archive's official repository (lukecavabarrett/pna) is linked to this paper; the 55 samples below come from 4 community repositories. Syntology harvested code from lukecavabarrett/pna for another paper.

lukecavabarrett/pna: Syntology's graph links these samples to Graph Attention Networks (arXiv:1710.10903; that paper's own run record: 61 ran (of which 28 constructed an object rather than computing a result; 52 with no instrument failure: 1 honoured, 2 violated, 49 with no contract checked; 9 where Syntology's instrument failed) · 45 unverified); nothing checks that they implement this paper's method.

Syntology 55 code samples harvested from 4 repositories linked to this paper: 33 ran (of which 4 constructed an object rather than computing a result; 11 with no instrument failure: 4 honoured, 0 violated, 7 with no contract checked; 22 where Syntology's instrument failed) · 22 unverified. Where Syntology's instrument failed: 17 where our contract draft was wrong; 5 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 55 samples from 4 repositories, 33 ran (11 with no instrument failure; 22 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 sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

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.

lukecavabarrett/pna officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Graph Attention Networks (arXiv:1710.10903; that paper's own run record: 61 ran (of which 28 constructed an object rather than computing a result; 52 with no instrument failure: 1 honoured, 2 violated, 49 with no contract checked; 9 where Syntology's instrument failed) · 45 unverified); nothing checks that they implement this paper's method. report
asarigun/GraphMixerNetworks mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to A Generalization of ViT/MLP-Mixer to Graphs (arXiv:2212.13350; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
changminwu/expandergnn mentioned on GitHubpytorchSyntology: 21 samples, 18 ran (11 instrument-failed) · 3 unverified report
cvignac/SMP mentioned on GitHubpytorchSyntology: 7 samples, 3 ran (1 instrument-failed) · 4 unverified report
dmlc/dgl mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to node2vec: Scalable Feature Learning for Networks (arXiv:1607.00653; that paper's own run record: 15 ran (of which 2 constructed an object rather than computing a result; 12 with no instrument failure: 2 honoured, 0 violated, 10 with no contract checked; 3 where Syntology's instrument failed) · 10 unverified), Interaction Networks for Learning about Objects, Relations and Physics (arXiv:1612.00222; that paper's own run record: 6 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 6 samples that ran constructed an object rather than computing a result), Junction Tree Variational Autoencoder for Molecular Graph Generation (arXiv:1802.04364; that paper's own run record: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result) and 3 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
rusty1s/pytorch_geometric mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to How Attentive are Graph Attention Networks? (arXiv:2105.14491; that paper's own run record: 11 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
Saro00/DGN pytorchSyntology: 26 samples, 12 ran (10 instrument-failed) · 14 unverified report
alexOarga/haiku-geometric jaxSyntology: 1 sample, 0 ran · 1 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

55 samples harvested: 33 ran (of which 4 constructed an object rather than computing a result; 11 with no instrument failure: 4 honoured, 0 violated, 7 with no contract checked; 22 where Syntology's instrument failed) · 22 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.

4ran · honoured contract
17ran · our draft was wrong
5ran; the contract check could not run
7ran
22unverified

Licence: 48 of the 55 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 4 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.

changminwu/expandergnn community (archive-listed) 21 samples, 18 ran (11 instrument-failed) · 3 unverified
ExpanderLinear layers/pna_layer.py ran no licence file found · pointer only · bbcfddab0672587b · report fixture: returned ExpanderLinear · level recorded 2026-09-22
ExpanderLinearFunction layers/pna_layer.py ran no licence file found · pointer only · 4aecc0c5ee5d5919 · report fixture: returned ExpanderLinearFunction · level recorded 2026-09-22
ExpanderScatterLinear layers/pna_layer.py ran no licence file found · pointer only · 528a64a346d98ccd · report fixture: returned ExpanderScatterLinear · level recorded 2026-09-22
LinearLayer in 14 papers layers/pna_layer.py ran no licence file found · pointer only · 6f4fac2426ac1ca3 · report fixture: returned LinearLayer · level recorded 2026-09-22
aggregate_max in 5 papers layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 0cb93154c335108f · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_mean in 9 papers layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · b19d1fdc25bad3bd · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_min in 5 papers layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · ff879af48e34176e · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_moment layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 481088b2fcfe7538 · report fixture: signature h:Tensor/2/float, n:int/0/int · returned Tensor · level recorded 2026-09-22
aggregate_moment_3 layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 12e2cedd3aa5d6ee · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_moment_4 layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 1459f6559164001b · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_moment_5 layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 4a7092a5a3c9f554 · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_std layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 6eb59d131c4ff569 · report fixture: signature h:Tensor/3/float · returned Tensor · level recorded 2026-09-22
aggregate_sum layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 5aeed6a78e303573 · report fixture: signature h:Tensor/2/float · returned Tensor · level recorded 2026-09-22
aggregate_var layers/pna_layer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 92fa1e926f5bd985 · report fixture: signature h:Tensor/3/float · returned Tensor · level recorded 2026-09-22
sampler in 7 papers layers/pna_layer.py ran · our draft was wrong no licence file found · pointer only · ea989e2ffeb38bb6 · report fixture: signature outdim:int/0/int, indim:int/0/int, density:float/0/float, method:str/0/str · returned tuple · level recorded 2026-09-22
scale_amplification layers/pna_layer.py ran · honoured contract no licence file found · pointer only · 5d07dbe669b9f73a · report fixture: signature h:float/0/float, D:int/0/int, avg_d:dict/0/? · returned float64 · level recorded 2026-09-22
scale_attenuation layers/pna_layer.py ran · honoured contract no licence file found · pointer only · d0d453f04b18e8f1 · report fixture: signature h:float/0/float, D:int/0/int, avg_d:dict/0/? · returned float64 · level recorded 2026-09-22
scale_identity layers/pna_layer.py ran · honoured contract fingerprinted no licence file found · pointer only · 2d32c33e16bad56e · report fixture: signature h:Tensor/2/float, D:NoneType/0/?, avg_d:NoneType/0/? · returned Tensor · level recorded 2026-09-22
MultiLinearLayer layers/pna_layer.py unverified no licence file found · pointer only · 71500fc15d19f2bc · report
PNALayer layers/pna_layer.py unverified no licence file found · pointer only · e8010b28fdc39a80 · report
PNATower layers/pna_layer.py unverified no licence file found · pointer only · 2c7376a202ec32d2 · report
cvignac/SMP community (archive-listed) 7 samples, 3 ran (1 instrument-failed) · 4 unverified
Linear in 127 papers models/smp_layers.py ran · metamorphic tier: invariant MIT (permissive) · e511d58ec99ede06 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
XtoX models/smp_layers.py ran · metamorphic tier: invariant MIT (permissive) · ef6906c0d3364ced · report fixture: signature x:Tensor/2/float, batch_info:NoneType/0/? · returned Tensor · level recorded 2026-09-22
kaiming_init_with_gain models/smp_layers.py ran · our draft was wrong MIT (permissive) · 244b5dd39c13f84a · report fixture: signature x:Tensor/2/float, gain:float/0/float, a:float/0/float, mode:str/0/str, nonlinearity:str/0/str · returned Tensor · level recorded 2026-09-22
EntryWiseX models/smp_layers.py unverified MIT (permissive) · 8e16971a959b54a3 · report
EntrywiseU models/smp_layers.py unverified MIT (permissive) · 9df9644a36246a29 · report
SMPLayer models/smp_layers.py unverified MIT (permissive) · 789af8973c8347ee · report
UtoU models/smp_layers.py unverified MIT (permissive) · 16f7d89050a5bb58 · report
Saro00/DGN community (archive-listed) 26 samples, 12 ran (10 instrument-failed) · 14 unverified
MLP in 375 papers models/pytorch/dgn_layer.py ran fingerprinted no licence file found · pointer only · 5fc1afe08991ab11 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-24
aggregate_identity models/pytorch/dgn_layer.py ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only · a03e9cc59c2671f7 · report fixture: signature X:Tensor/4/float, adj:Tensor/2/float, eigvec:NoneType/0/?, self_loop:bool/0/bool, device:str/0/str, avg_d:NoneType/0/? · returned Tensor · level recorded 2026-09-22
aggregate_max in 5 papers models/pytorch/dgn_layer.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 6a8ff41179e540bd · report fixture: signature X:Tensor/4/float, adj:Tensor/3/float, eigvec:Tensor/2/float, min_value:float/0/float, self_loop:bool/0/bool, device:str/0/str, avg_d:NoneType/0/? · returned Tensor · level recorded 2026-09-22
aggregate_min in 5 papers models/pytorch/dgn_layer.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 3b1fc26527b0808f · report fixture: signature X:Tensor/4/float, adj:Tensor/3/float, eigvec:NoneType/0/?, max_value:float/0/float, self_loop:bool/0/bool, device:str/0/str, avg_d:NoneType/0/? · returned Tensor · level recorded 2026-09-22
aggregate_normalised_mean models/pytorch/dgn_layer.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 7e7c9a7e4e3f56ed · report fixture: signature X:Tensor/4/float, adj:Tensor/3/float, eigvec:NoneType/0/?, self_loop:bool/0/bool, device:str/0/str, avg_d:NoneType/0/? · returned Tensor · level recorded 2026-09-22
get_activation in 218 papers models/pytorch/dgn_layer.py ran · our draft was wrong no licence file found · pointer only · fe76e55f0b804b7b · report fixture: signature activation:str/0/str · returned ReLU · level recorded 2026-09-22
get_adjacency_from_gradient_of_features models/pytorch/dgn_layer.py ran · our draft was wrong no licence file found · pointer only · 00473157272f2713 · report fixture: signature adj:Tensor/3/float, features:Tensor/2/float, normalization:str/0/str, add_diag:bool/0/bool, absolute_adj:bool/0/bool · returned Tensor · level recorded 2026-09-22
get_eig_adjacency models/pytorch/dgn_layer.py ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only · b4f4cbb16613e6d5 · report fixture: signature adj:Tensor/3/float, eig_idx:list/1/int, eigvec:Tensor/3/float, normalization:str/0/str, add_diag:bool/0/bool, absolute_adj:bool/0/bool, normalize_L:bool/0/bool, eig_acos:bool/0/bool · returned dict · level recorded 2026-09-22
scale_amplification models/pytorch/dgn_layer.py ran · our draft was wrong no licence file found · pointer only · 1d260f12223b0d23 · report fixture: signature X:Tensor/2/float, adj:Tensor/2/float, avg_d:dict/0/? · returned Tensor · level recorded 2026-09-22
scale_attenuation models/pytorch/dgn_layer.py ran · our draft was wrong no licence file found · pointer only · e1422930e833ac1b · report fixture: signature X:Tensor/2/float, adj:Tensor/2/float, avg_d:dict/0/? · returned Tensor · level recorded 2026-09-22
scale_identity models/pytorch/dgn_layer.py ran · honoured contract fingerprinted no licence file found · pointer only · 3d28677798f20e1f · report fixture: signature X:Tensor/2/float, adj:Tensor/2/float, avg_d:Tensor/2/float · returned Tensor · level recorded 2026-09-22
scale_linear models/pytorch/dgn_layer.py ran · our draft was wrong no licence file found · pointer only · b673916e3a39dd5d · report fixture: signature X:Tensor/2/float, adj:Tensor/2/float, avg_d:dict/0/? · returned Tensor · level recorded 2026-09-22
DGNLayer models/pytorch/dgn_layer.py unverified no licence file found · pointer only · e4b24e60ca0d9599 · report
DGNTower models/pytorch/dgn_layer.py unverified no licence file found · pointer only · ce5e689ba09ce45d · report
FCLayer models/pytorch/dgn_layer.py unverified no licence file found · pointer only · 3aa8e705794c9d66 · report
aggregate_dirs models/pytorch/dgn_layer.py unverified no licence file found · pointer only · aeafec5c97bd497f · report
aggregate_mean in 9 papers models/pytorch/dgn_layer.py unverified no licence file found · pointer only · d7ead85a1d37c169 · report
aggregate_mean_amplified models/pytorch/dgn_layer.py unverified no licence file found · pointer only · be921ec112433f72 · report
aggregate_mean_attenuated models/pytorch/dgn_layer.py unverified no licence file found · pointer only · 89ec98a7ccf842eb · report
aggregate_moment_rooted models/pytorch/dgn_layer.py unverified no licence file found · pointer only · cfb905cff1a241fd · report
aggregate_softmax models/pytorch/dgn_layer.py unverified no licence file found · pointer only · 94492fb6844ac8f1 · report
aggregate_softmin models/pytorch/dgn_layer.py unverified no licence file found · pointer only · 3d7c8bb83a9f2a81 · report
aggregate_std models/pytorch/dgn_layer.py unverified no licence file found · pointer only · d38bd6844f4c5c7e · report
aggregate_sum models/pytorch/dgn_layer.py unverified no licence file found · pointer only · c6d0524ad9a473ce · report
aggregate_var models/pytorch/dgn_layer.py unverified no licence file found · pointer only · 1ad750797e80563a · report
scale_inverse_linear models/pytorch/dgn_layer.py unverified no licence file found · pointer only · c53e8d8bb36479c1 · report
alexOarga/haiku-geometric community (archive-listed) 1 sample, 0 ran · 1 unverified
PNAConv haiku_geometric/nn/conv/pna_conv.py unverified no licence file found · pointer only · f610d8297c271f3c · 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. 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. 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.

4 assemblies on this paper: 4 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

4 sibling assemblies for Principal Neighbourhood Aggregation (PNA) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Principal Neighbourhood Aggregation (PNA) (graph method: Principal Neighbourhood Aggregation (PNA)). 26 bricks; individually on a fixture: 12 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 10 where Syntology's instrument failed) · 14 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. DGNLayer entry · level 0 · unverified · e4b24e60ca0d9599
  2. DGNTower helper · level 0 · unverified · ce5e689ba09ce45d
  3. FCLayer helper · level 0 · unverified · 3aa8e705794c9d66
  4. MLP helper · level 1 · ran · 5fc1afe08991ab11
  5. get_activation helper · level 1 · ran · our draft was wrong · fe76e55f0b804b7b
  6. aggregate_dirs helper · level 0 · unverified · aeafec5c97bd497f
  7. aggregate_identity helper · level 1 · ran; the contract check could not run (RAISES) · a03e9cc59c2671f7
  8. aggregate_max helper · level 1 · ran; the contract check could not run (RAISES) · 6a8ff41179e540bd
  9. aggregate_mean helper · level 0 · unverified · d7ead85a1d37c169
  10. aggregate_mean_amplified helper · level 0 · unverified · be921ec112433f72
  11. aggregate_mean_attenuated helper · level 0 · unverified · 89ec98a7ccf842eb
  12. aggregate_min helper · level 1 · ran; the contract check could not run (RAISES) · 3b1fc26527b0808f
  13. aggregate_moment_rooted helper · level 0 · unverified · cfb905cff1a241fd
  14. aggregate_normalised_mean helper · level 1 · ran; the contract check could not run (RAISES) · 7e7c9a7e4e3f56ed
  15. aggregate_softmax helper · level 0 · unverified · 94492fb6844ac8f1
  16. aggregate_softmin helper · level 0 · unverified · 3d7c8bb83a9f2a81
  17. aggregate_std helper · level 0 · unverified · d38bd6844f4c5c7e
  18. aggregate_sum helper · level 0 · unverified · c6d0524ad9a473ce
  19. aggregate_var helper · level 0 · unverified · 1ad750797e80563a
  20. get_adjacency_from_gradient_of_features helper · level 1 · ran · our draft was wrong · 00473157272f2713
  21. get_eig_adjacency helper · level 1 · ran; the contract check could not run (RAISES) · b4f4cbb16613e6d5
  22. scale_amplification helper · level 1 · ran · our draft was wrong · 1d260f12223b0d23
  23. scale_attenuation helper · level 1 · ran · our draft was wrong · e1422930e833ac1b
  24. scale_identity helper · level 1 · ran · honoured contract · 3d28677798f20e1f
  25. scale_inverse_linear helper · level 0 · unverified · c53e8d8bb36479c1
  26. scale_linear helper · level 1 · ran · our draft was wrong · b673916e3a39dd5d

harvested Assembly for Principal Neighbourhood Aggregation (PNA) (graph method: Principal Neighbourhood Aggregation (PNA)). 21 bricks; individually on a fixture: 18 ran (of which 4 constructed an object rather than computing a result; 7 with no instrument failure: 3 honoured, 0 violated, 4 with no contract checked; 11 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. PNALayer entry · level 0 · unverified · e8010b28fdc39a80
  2. LinearLayer helper · level 1 · ran · 6f4fac2426ac1ca3
  3. PNATower model · level 0 · unverified · 2c7376a202ec32d2
  4. ExpanderLinear helper · level 1 · ran · bbcfddab0672587b
  5. ExpanderScatterLinear helper · level 1 · ran · 528a64a346d98ccd
  6. MultiLinearLayer helper · level 0 · unverified · 71500fc15d19f2bc
  7. ExpanderLinearFunction helper · level 1 · ran · 4aecc0c5ee5d5919
  8. sampler helper · level 1 · ran · our draft was wrong · ea989e2ffeb38bb6
  9. aggregate_max helper · level 1 · ran · our draft was wrong · 0cb93154c335108f
  10. aggregate_mean helper · level 1 · ran · our draft was wrong · b19d1fdc25bad3bd
  11. aggregate_min helper · level 1 · ran · our draft was wrong · ff879af48e34176e
  12. aggregate_moment helper · level 1 · ran · our draft was wrong · 481088b2fcfe7538
  13. aggregate_moment_3 helper · level 1 · ran · our draft was wrong · 12e2cedd3aa5d6ee
  14. aggregate_moment_4 helper · level 1 · ran · our draft was wrong · 1459f6559164001b
  15. aggregate_moment_5 helper · level 1 · ran · our draft was wrong · 4a7092a5a3c9f554
  16. aggregate_std helper · level 1 · ran · our draft was wrong · 6eb59d131c4ff569
  17. aggregate_sum helper · level 1 · ran · our draft was wrong · 5aeed6a78e303573
  18. aggregate_var helper · level 1 · ran · our draft was wrong · 92fa1e926f5bd985
  19. scale_amplification helper · level 1 · ran · honoured contract · 5d07dbe669b9f73a
  20. scale_attenuation helper · level 1 · ran · honoured contract · d0d453f04b18e8f1
  21. scale_identity helper · level 1 · ran · honoured contract · 2d32c33e16bad56e

harvested Assembly for Principal Neighbourhood Aggregation (PNA) (graph method: Principal Neighbourhood Aggregation (PNA)). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. PNAConv entry · level 0 · unverified · f610d8297c271f3c

harvested Assembly for Principal Neighbourhood Aggregation (PNA) (graph method: Principal Neighbourhood Aggregation (PNA)). 7 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. SMPLayer entry · level 0 · unverified · 789af8973c8347ee
  2. EntryWiseX helper · level 0 · unverified · 8e16971a959b54a3
  3. EntrywiseU helper · level 0 · unverified · 9df9644a36246a29
  4. UtoU helper · level 0 · unverified · 16f7d89050a5bb58
  5. XtoX helper · level 1 · ran · ef6906c0d3364ced
  6. Linear helper · level 1 · ran · e511d58ec99ede06
  7. kaiming_init_with_gain helper · level 1 · ran · our draft was wrong · 244b5dd39c13f84a

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="2004.05718") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Graph ClassificationGraph RegressionMolecular Property PredictionNode Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Graph Classification CIFAR10 100k PNA Accuracy (%) 70.47 #12 of 20 Archive leaderboard report
Graph Classification D&D PNA Accuracy 78.992±4.407 #19 of 53 Archive leaderboard report
Graph Classification ENZYMES PNA Accuracy 73.021±2.512 #9 of 54 Archive leaderboard report
Graph Classification IMDb-B PNA Accuracy 78.000±3.808 #13 of 51 Archive leaderboard report
Graph Classification NCI1 PNA Accuracy 84.964±1.391 #16 of 69 Archive leaderboard report
Graph Classification NCI109 PNA Accuracy 83.382±1.045 #12 of 38 Archive leaderboard report
Graph Classification PROTEINS PNA Accuracy 77.679±3.281 #29 of 103 Archive leaderboard report
Graph Property Prediction ogbg-molhiv PNA Ext. data No #25 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv PNA Number of params 326081 #25 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv PNA Test ROC-AUC 0.7905 ± 0.0132 #25 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv PNA Validation ROC-AUC 0.8519 ± 0.0099 #25 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molpcba PNA Ext. data No #24 of 36 Archive leaderboard report
Graph Property Prediction ogbg-molpcba PNA Number of params 6550839 #24 of 36 Archive leaderboard report
Graph Property Prediction ogbg-molpcba PNA Test AP 0.2838 ± 0.0035 #24 of 36 Archive leaderboard report
Graph Property Prediction ogbg-molpcba PNA Validation AP 0.2926 ± 0.0026 #24 of 36 Archive leaderboard report
Graph Regression ESR2 PNA R2 0.696±0.000 #2 of 9 Archive leaderboard report
Graph Regression ESR2 PNA RMSE 0.486±0.696 #2 of 9 Archive leaderboard report
Graph Regression F2 PNA R2 0.891±0.000 #2 of 9 Archive leaderboard report
Graph Regression F2 PNA RMSE 0.336±0.891 #2 of 9 Archive leaderboard report
Graph Regression KIT PNA R2 0.843±0.000 #1 of 9 Archive leaderboard report
Graph Regression KIT PNA RMSE 0.430±0.843 #1 of 9 Archive leaderboard report
Graph Regression Lipophilicity PNA R2 0.830±0.007 #1 of 23 Archive leaderboard report
Graph Regression Lipophilicity PNA RMSE 0.520±0.011 #1 of 23 Archive leaderboard report
Graph Regression PARP1 PNA R2 0.924±0.000 #2 of 9 Archive leaderboard report
Graph Regression PARP1 PNA RMSE 0.346±0.924 #2 of 9 Archive leaderboard report
Graph Regression PGR PNA R2 0.717±0.000 #2 of 9 Archive leaderboard report
Graph Regression PGR PNA RMSE 0.514±0.717 #2 of 9 Archive leaderboard report
Graph Regression ZINC PNA MAE 0.142 #22 of 27 Archive leaderboard report
Graph Regression ZINC-full PNA Test MAE 0.057±0.007 #14 of 19 Archive leaderboard report
Molecular Property Prediction ESOL PNA R2 0.942±0.006 #2 of 20 Archive leaderboard report
Molecular Property Prediction ESOL PNA RMSE 0.493±0.026 #2 of 20 Archive leaderboard report
Molecular Property Prediction FreeSolv PNA R2 0.951±0.009 #7 of 22 Archive leaderboard report
Molecular Property Prediction FreeSolv PNA RMSE 0.870±0.081 #7 of 22 Archive leaderboard report
Node Classification PATTERN 100k PNA Accuracy (%) 86.567 #4 of 9 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

Introduced by this paper: PNA

PNA

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