Papers › Rethinking Neural Operations for Diverse Tasks

Rethinking Neural Operations for Diverse Tasks

29 Mar 2021NeurIPS 2021 12arXiv:2103.15798archive 2025-07-28

Nicholas Roberts, Mikhail Khodak, Tri Dao, Liam Li, Christopher Ré, Ameet Talwalkar

An important goal of AutoML is to automate-away the design of neural networks on new tasks in under-explored domains. Motivated by this goal, we study the problem of enabling users to discover the right neural operations given data from their specific domain. We introduce a search space of operations called XD-Operations that mimic the inductive bias of standard multi-channel convolutions while being much more expressive: we prove that it includes many named operations across multiple application areas. Starting with any standard backbone such as ResNet, we show how to transform it into a search space over XD-operations and how to traverse the space using a simple weight-sharing scheme. On a diverse set of tasks -- solving PDEs, distance prediction for protein folding, and music modeling -- our approach consistently yields models with lower error than baseline networks and often even lower error than expert-designed domain-specific approaches.

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

Syntology 34 code samples harvested from 3 repositories linked to this paper: 22 ran (of which 8 constructed an object rather than computing a result; 17 with no instrument failure: 3 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 12 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 33 samples from 2 repositories, 22 ran (17 with no instrument failure; 5 where Syntology's instrument failed); found in paper text by Syntology: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 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 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 5 of the 22 samples that ran have no recorded date.

mkhodak/relax officialmentioned in papermentioned on GitHubpytorchSyntology: 24 samples, 15 ran (2 instrument-failed) · 9 unverified report
nick11roberts/XD officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 9 samples, 7 ran (3 instrument-failed) · 2 unverified report
sjunhongshen/dash mentioned on GitHubpytorchSyntology: not harvested 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

34 samples harvested: 22 ran (of which 8 constructed an object rather than computing a result; 17 with no instrument failure: 3 honoured, 0 violated, 14 with no contract checked; 5 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 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.

3ran · honoured contract
4ran · our draft was wrong
1ran; the contract check could not run
14ran
12unverified

Licence: 1 of the 34 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 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”. For 7 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

mkhodak/relax official repository 24 samples, 15 ran (2 instrument-failed) · 9 unverified
AvgPool relax/xd.py ran · our draft was wrong MIT (permissive) · 2a31a6dcb06ddf63 · report fixture: signature dims:int/0/int · returned type · level recorded 2026-09-22
Complex2Real relax/xd.py ran · metamorphic tier: well formed fingerprinted MIT (permissive) · bec5cac219d82ded · report fixture: signature x:Tensor/2/complex · returned Tensor · level recorded 2026-09-22
ConvPerm relax/xd.py ran MIT (permissive) · ba89e07cd33d1453 · report fixture: returned ConvPerm · level recorded 2026-09-22
FNOPerm relax/xd.py ran fingerprinted MIT (permissive) · 746f20753fb4022f · report fixture: returned FNOPerm · level recorded 2026-09-22
Fourier relax/xd.py ran fingerprinted MIT (permissive) · 4b400ff00a487d94 · report fixture: returned Fourier · level recorded 2026-09-22
Pad relax/xd.py ran fingerprinted MIT (permissive) · 1dde65deadfc1f2d · report fixture: returned Pad · level recorded 2026-09-22
TransposeFlip relax/xd.py ran fingerprinted MIT (permissive) · 7113106c9caa90dc · report fixture: returned TransposeFlip · level recorded 2026-09-22
TransposePad relax/xd.py ran fingerprinted MIT (permissive) · 63dc4a7a7bfdd3ba · report fixture: returned TransposePad · level recorded 2026-09-22
TransposePerm relax/xd.py ran fingerprinted MIT (permissive) · 6817170fe598f430 · report fixture: returned TransposePerm · level recorded 2026-09-22
TruncateHermitian relax/xd.py ran fingerprinted MIT (permissive) · df99f62038dc6b95 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
Unpad relax/xd.py ran MIT (permissive) · 83fd4dada951a4ce · report fixture: returned Unpad · level recorded 2026-09-22
atrous_permutation relax/xd.py ran · honoured contract fingerprinted MIT (permissive) · 8597cc597b0fd17f · report fixture: signature n:int/0/int, k:int/0/int, d:int/0/int · returned Tensor · level recorded, date not recorded
int2tuple relax/xd.py ran · honoured contract fingerprinted MIT (permissive) · b65e5411454bdb5b · report fixture: signature int_or_tuple:int/0/int, length:int/0/int, allow_other:bool/0/bool · returned tuple · level recorded 2026-09-22
transpose_permutation relax/xd.py ran · honoured contract fingerprinted MIT (permissive) · c4904be3964ec72b · report fixture: signature n:int/0/int, k:int/0/int, d:int/0/int · returned Tensor · level recorded, date not recorded
truncate_freq relax/xd.py ran · our draft was wrong MIT (permissive) · 7378c3d22e68d903 · report fixture: signature in_size:list/1/int, freqs:list/1/int · returned generator · level recorded 2026-09-22
SharedOperation relax/xd.py unverified MIT (permissive) · 94357aa818c939d9 · report
TensorProduct relax/xd.py unverified MIT (permissive) · da15dcc8176fc0a6 · report
XD relax/xd.py unverified MIT (permissive) · 0a723f7b3a1f7560 · report
fixed relax/xd.py unverified MIT (permissive) · 50254999b8752976 · report
multichannel_prod relax/xd.py unverified MIT (permissive) · 50200b49d8c95110 · report
pad2size relax/xd.py unverified MIT (permissive) · bc0b1e3f81223bd5 · report
resnet20 in 42 papers examples/resnet/resnet.py unverified MIT (permissive) · b3920efd145667ac · report
resnet32 in 47 papers examples/resnet/resnet.py unverified MIT (permissive) · ccbabd3d954fa0a8 · report
resnet44 in 19 papers examples/resnet/resnet.py unverified MIT (permissive) · 595fe4a7995fd021 · report
nick11roberts/XD official repository 9 samples, 7 ran (3 instrument-failed) · 2 unverified
MaxPool chrysalis/darts.py ran Apache-2.0 (permissive) · 1ff4c96878cdae81 · report fixture: signature dims:int/0/int · returned type · level recorded 2026-09-25
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
DARTS/utils.py ran Apache-2.0 (permissive) · e93d9ddbfeb73ced · report fixture: signature output:Tensor/2/float, target:Tensor/1/int, topk:tuple/1/int · returned list · level recorded 2026-09-25
check_weight_norm chrysalis/chrysalis.py ran Apache-2.0 (permissive) · 2480ed0be135e57a · report fixture: signature module:callable/0/? · returned WeightNorm · level recorded 2026-09-25
complex_dist chrysalis/xd.py ran · our draft was wrong Apache-2.0 (permissive) · 9ec5d46f798ca92d · report fixture: signature X:Tensor/2/complex, Y:Tensor/2/complex, relative:bool/0/bool, scaled:bool/0/bool · returned Tensor · level recorded, date not recorded
complex_norm chrysalis/xd.py ran · our draft was wrong Apache-2.0 (permissive) · f99717b5f5887edd · report fixture: signature X:Tensor/2/complex · returned Tensor · level recorded, date not recorded
get_module in 37 papers chrysalis/chrysalis.py ran Apache-2.0 (permissive) · e151df3b5424a3fb · report fixture: signature model:callable/0/?, module_string:str/0/str · returned Linear · level recorded 2026-09-25
special_dist chrysalis/xd.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · cc788ee6ee06b4df · report fixture: signature X:Tensor/2/float, Y:Tensor/2/float, relative:bool/0/bool, scaled:bool/0/bool · returned Tensor · level recorded, date not recorded
count_parameters_in_MB in 14 papers DARTS/utils.py unverified Apache-2.0 (permissive) · 3e816289579cc9e9 · report
drop_path in 325 papers
census: 27 implementations from 61 papers, 5 distinct outputs across 2 buckets, one shared input per bucket
DARTS/utils.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 13c2719404e25b25 · report
ba-lab/pdnet found in paper text by Syntology 1 sample, 0 ran · 1 unverified
basic_fcn models.py unverified Apache-2.0 (permissive) · 68a0988ac016a394 · 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 XD-Operations (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for XD-Operations (graph method: XD-Operations). 21 bricks; individually on a fixture: 15 ran (of which 8 constructed an object rather than computing a result; 13 with no instrument failure: 3 honoured, 0 violated, 10 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. XD entry · level 0 · unverified · 0a723f7b3a1f7560
  2. SharedOperation helper · level 0 · unverified · 94357aa818c939d9
  3. fixed helper · level 0 · unverified · 50254999b8752976
  4. int2tuple helper · level 1 · ran · honoured contract · b65e5411454bdb5b
  5. multichannel_prod helper · level 0 · unverified · 50200b49d8c95110
  6. pad2size helper · level 0 · unverified · bc0b1e3f81223bd5
  7. truncate_freq helper · level 1 · ran · our draft was wrong · 7378c3d22e68d903
  8. Complex2Real helper · level 1 · ran · bec5cac219d82ded
  9. ConvPerm helper · level 1 · ran · ba89e07cd33d1453
  10. FNOPerm helper · level 1 · ran · 746f20753fb4022f
  11. Fourier helper · level 1 · ran · 4b400ff00a487d94
  12. Pad helper · level 1 · ran · 1dde65deadfc1f2d
  13. TensorProduct helper · level 0 · unverified · da15dcc8176fc0a6
  14. TransposeFlip helper · level 1 · ran · 7113106c9caa90dc
  15. TransposePad helper · level 1 · ran · 63dc4a7a7bfdd3ba
  16. TransposePerm helper · level 1 · ran · 6817170fe598f430
  17. TruncateHermitian helper · level 1 · ran · df99f62038dc6b95
  18. Unpad helper · level 1 · ran · 83fd4dada951a4ce
  19. atrous_permutation helper · level 1 · ran · honoured contract · 8597cc597b0fd17f
  20. transpose_permutation helper · level 1 · ran · honoured contract · c4904be3964ec72b
  21. AvgPool helper · level 1 · ran · our draft was wrong · 2a31a6dcb06ddf63

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

one of 2 siblings

  1. basic_fcn entry · level 0 · unverified · 68a0988ac016a394

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="2103.15798") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

AutoMLImage ClassificationInductive BiasMusic ModelingNeural Architecture SearchProtein Folding

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingKaiming InitializationMax PoolingReLUResidual BlockResidual Connection

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