Papers › Relational inductive biases, deep learning, and graph networks
Relational inductive biases, deep learning, and graph networks
Peter W. Battaglia, Jessica B. Hamrick, Victor Bapst, Alvaro Sanchez-Gonzalez, Vinicius Zambaldi, Mateusz Malinowski, Andrea Tacchetti, David Raposo, Adam Santoro, Ryan Faulkner, Caglar Gulcehre, Francis Song, Andrew Ballard, Justin Gilmer, George Dahl, Ashish Vaswani, Kelsey Allen, Charles Nash, Victoria Langston, Chris Dyer, Nicolas Heess, Daan Wierstra, Pushmeet Kohli, Matt Botvinick, Oriol Vinyals, Yujia Li, Razvan Pascanu
Artificial intelligence (AI) has undergone a renaissance recently, making major progress in key domains such as vision, language, control, and decision-making. This has been due, in part, to cheap data and cheap compute resources, which have fit the natural strengths of deep learning. However, many defining characteristics of human intelligence, which developed under much different pressures, remain out of reach for current approaches. In particular, generalizing beyond one's experiences--a hallmark of human intelligence from infancy--remains a formidable challenge for modern AI. The following is part position paper, part review, and part unification. We argue that combinatorial generalization must be a top priority for AI to achieve human-like abilities, and that structured representations and computations are key to realizing this objective. Just as biology uses nature and nurture cooperatively, we reject the false choice between "hand-engineering" and "end-to-end" learning, and instead advocate for an approach which benefits from their complementary strengths. We explore how using relational inductive biases within deep learning architectures can facilitate learning about entities, relations, and rules for composing them. We present a new building block for the AI toolkit with a strong relational inductive bias--the graph network--which generalizes and extends various approaches for neural networks that operate on graphs, and provides a straightforward interface for manipulating structured knowledge and producing structured behaviors. We discuss how graph networks can support relational reasoning and combinatorial generalization, laying the foundation for more sophisticated, interpretable, and flexible patterns of reasoning. As a companion to this paper, we have released an open-source software library for building graph networks, with demonstrations of how to use them in practice.
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):
get_harvested_code_for_paper(arxiv_id="1806.01261")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="1806.01261", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="1806.01261")
Code
The archive's official repository (deepmind/graph_nets) was not harvested; the 51 samples below come from 7 community repositories.
Syntology 51 code samples harvested from 7 repositories linked to this paper: 39 ran (of which 5 constructed an object rather than computing a result; 38 with no instrument failure: 4 honoured, 0 violated, 34 with no contract checked; 1 where Syntology's instrument failed) · 12 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.
By repository: community (archive-listed): 51 samples from 7 repositories, 39 ran (38 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 sample is linked to this paper from 6 of its repositories that Syntology harvested for other papers:
- LucaHermes/graph-unet-traffic-prediction, community (archive-listed): harvested for another paper, listed under the repository
- materialsvirtuallab/matgl, community (archive-listed): harvested for another paper, listed under the repository
- materialsvirtuallab/megnet, community (archive-listed): harvested for another paper, listed under the repository
- nrel/m2p, community (archive-listed): harvested for another paper, listed under the repository
- patrickhart/jaxdl, community (archive-listed): harvested for another paper, listed under the repository
- raphaelavalos/attention_tsp_graph_net, community (archive-listed): harvested for another paper, listed under the repository
Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.
31 repositories listed; official and paper-mentioned ones first.
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
51 samples harvested: 39 ran (of which 5 constructed an object rather than computing a result; 38 with no instrument failure: 4 honoured, 0 violated, 34 with no contract checked; 1 where Syntology's instrument failed) · 12 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.
Licence: 15 of the 51 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 7 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.
bab3c705163c083f · report
fixture: signature n:int/0/int, d:float/0/float · returned dict · level recorded 2026-09-26 41b325265ab294e1 · report
fixture: signature nodes:Tensor/2/float, force_per_node:Tensor/2/float, step_size:float/0/float · returned Tensor · level recorded 2026-09-26 6c8e3bafab6c39b9 · report
fixture: signature num_nodes:int/0/int, num_edges:int/0/int · returned dict · level recorded 2026-09-26 c843d356064d8658 · report
fixture: signature receiver_nodes:SymbolicTensor/2/float, sender_nodes:SymbolicTensor/2/float, k:SymbolicTensor/1/float, x_rest:SymbolicTensor/1/float · returned EagerTensor · level recorded 2026-09-26 3eb164551e02b4d4 · report
fixture: signature graph:DiGraph/0/?, ax:Axes/0/?, pos:NoneType/0/? · returned dict · level recorded 2026-09-26 33a8530945e1ca58 · report
fixture: returned Aggregator · level recorded 2026-09-22 7ff65c09e83a289d · report
fixture: returned EdgeBlock · level recorded 2026-09-22 2449c07f88cb5df0 · report
fixture: returned GlobalBlock · level recorded 2026-09-22 71e81adbbb88a6bd · report
fixture: returned GraphsTuple · level recorded 2026-09-22 023b96c5f1ca64d0 · report
fixture: returned NodeBlock · level recorded 2026-09-22 050689c96b6e52e3 · report
fixture: signature graph:GraphsTuple/1/Tensor · returned Tensor · level recorded 2026-09-22 84ee378d07f846fe · report
fixture: signature graph:GraphsTuple/1/Tensor · returned Tensor · level recorded 2026-09-22 049b1bb616f6d97d · report
fixture: signature graph:GraphsTuple/1/Tensor · returned Tensor · level recorded 2026-09-22 f296fbbdafe6167f · report
fixture: signature graph:GraphsTuple/1/Tensor · returned Tensor · level recorded 2026-09-22 690a776c071f826a · report
eca4d5c0e04f01b2 · report
c79feac3c63518bf · report
5bbc9573fbe03c59 · report
c4aeeb542f565ba2 · report
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket megnet/utils/metrics.py ran fingerprinted BSD-3-Clause (permissive) ·
793d95f974b73be9 · report
fixture: signature y_true:ndarray/1/int, y_pred:ndarray/1/float · returned float64 · level recorded 2026-09-25 4fc6585b7b5fce76 · report
fixture: signature x:ndarray/3/float · returned ndarray · level recorded 2026-09-25 88f0e66d192714e6 · report
fixture: signature value:str/0/str, labels:list/1/str · returned list · level recorded 2026-09-26 census: 4 implementations from 6 papers, 3 distinct outputs across 3 buckets, one shared input per bucket megnet/utils/metrics.py ran fingerprinted BSD-3-Clause (permissive) ·
7cd70353cde137d0 · report
fixture: signature y_true:ndarray/1/float, y_pred:ndarray/1/float · returned float · level recorded 2026-09-25 310cd70b817b692c · report
fixture: signature y_true:EagerTensor/2/float, y_pred:EagerTensor/2/float, scale:int/0/int · returned SymbolicTensor · level recorded 2026-09-25 2d4a31854d09c4ab · report
fixture: signature x:EagerTensor/3/float, index:EagerTensor/1/int, axis:int/0/int · returned EagerTensor · level recorded 2026-09-26 b72502cbc4652705 · report
fixture: returned EagerTensor · level recorded 2026-09-25 083fdc1346d4a41f · report
fixture: signature embed_edge_fn:callable/0/?, embed_node_fn:callable/0/?, embed_global_fn:callable/0/? · returned function · level recorded 2026-09-24 7c803bc1dbe9c38f · report
fixture: signature latent_size:int/0/int · returned function · level recorded 2026-09-26 6c720ecf7652656d · report
fixture: signature data:ArrayImpl/1/float, segment_ids:ArrayImpl/1/int, num_segments:int/0/int, indices_are_sorted:bool/0/bool, unique_indices:bool/0/bool · returned ArrayImpl · level recorded 2026-09-24 b02dcff8929a613b · report
fixture: signature data:ArrayImpl/1/float, segment_ids:ArrayImpl/1/int, num_segments:int/0/int, indices_are_sorted:bool/0/bool, unique_indices:bool/0/bool · returned ArrayImpl · level recorded 2026-09-24 fa85b4d47d64d575 · report
fixture: signature data:ArrayImpl/1/float, segment_ids:ArrayImpl/1/int, num_segments:int/0/int, indices_are_sorted:bool/0/bool, unique_indices:bool/0/bool · returned ArrayImpl · level recorded 2026-09-24 67bfb69a00d165f3 · report
fixture: signature instances:list/1/SimpleCNF · returned BatchCNF · level recorded 2026-09-26 c78d05675edb4799 · report
fixture: signature M:Tensor/2/float, check:callable/0/?, val:callable/0/? · returned tuple · level recorded 2026-09-26 795cf0cae79b3a4c · report
fixture: signature n:int/0/int, m:int/0/int, M:tuple/1/list · returned zip · level recorded 2026-09-26 census: 36 implementations from 56 papers, 7 distinct outputs across 3 buckets, one shared input per bucket neurosat/logutil.py ran fingerprinted MIT (permissive) ·
c6a8334164beb95f · report
fixture: signature x:ndarray/1/float, derivative:bool/0/bool · returned ndarray · level recorded 2026-09-26 7994f503ff3f8886 · report
fixture: signature M_sparse:tuple/1/list, default:float/0/float · returned ndarray · level recorded 2026-09-26 9391c2ef8916602d · report
fixture: signature n:int/0/int, m:int/0/int, F:list/1/list · returned ndarray · level recorded 2026-09-26 b433f1f94987c049 · report
fixture: signature xs:list/1/ndarray, axis:int/0/int · returned EagerTensor · level recorded 2026-09-26 deebc52fd00d5008 · report
fixture: signature x:ndarray/2/float · returned int · level recorded 2026-09-26 0064a27d6d3cb322 · report
fixture: signature params:EagerTensor/2/float, indices:ndarray/1/int, axis:int/0/int · returned EagerTensor · level recorded 2026-09-26 1bc1b5ac372ed56e · report
fixture: signature _args:Namespace/0/? · returned int · level recorded 2026-09-26 e9c5142e7cbd84c4 · report
fixture: signature root:str/0/str · returned tuple · level recorded 2026-09-26 489a7393def35150 · report
fixture: signature units:list/1/int, input_tensor_list:list/1/KerasTensor, output_shape:tuple/1/int, activation:str/0/str, activate_last_layer:bool/0/bool, layernorm_last_layer:bool/0/bool, model_name:str/0/str · returned Functional · level recorded 2026-09-26 444ab2a8d8818a53 · report
fixture: signature step_times_ms:list/1/float · returned dict · level recorded 2026-09-26 77730fffb44065a5 · report
acc81d203eb667dc · report
bba2920bf9e6ce2d · report
30dd2468e72fd7e4 · report
9ee2d09dfb917e7d · 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.
2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).
2 sibling assemblies for graph network (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.
harvested Assembly for graph network (graph method: graph network). 10 bricks; individually on a fixture: 9 ran (of which 5 constructed an object rather than computing a result; 9 with no instrument failure: 4 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 2 siblings
- GraphNetwork entry · level 0 · unverified ·
6fd107ff14ac86bd - EdgeBlock model · level 1 · ran ·
7ff65c09e83a289d - GlobalBlock model · level 1 · ran ·
2449c07f88cb5df0 - NodeBlock model · level 1 · ran ·
023b96c5f1ca64d0 - GraphsTuple helper · level 1 · ran ·
71e81adbbb88a6bd - broadcast_globals_to_edges helper · level 1 · ran · honoured contract ·
050689c96b6e52e3 - broadcast_receiver_nodes_to_edges helper · level 1 · ran · honoured contract ·
049b1bb616f6d97d - broadcast_sender_nodes_to_edges helper · level 1 · ran · honoured contract ·
f296fbbdafe6167f - Aggregator helper · level 1 · ran ·
33a8530945e1ca58 - broadcast_globals_to_nodes helper · level 1 · ran · honoured contract ·
84ee378d07f846fe
harvested Assembly for graph network (graph method: graph network). 5 bricks; individually on a fixture: 0 ran · 5 unverified. The assembly has not been run end-to-end.
one of 2 siblings
- GraphNetwork entry · level 0 · unverified ·
5bbc9573fbe03c59 - EdgeModel model · level 0 · unverified ·
eca4d5c0e04f01b2 - GlobalModel model · level 0 · unverified ·
c79feac3c63518bf - NodeModel model · level 0 · unverified ·
c4aeeb542f565ba2 - BaseModel helper · level 0 · unverified ·
690a776c071f826a
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="1806.01261") returns these assemblies with each brick's role and wiring (how to connect).
Tasks
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