Papers › Efficient Neural Architecture Search via Parameter Sharing

Efficient Neural Architecture Search via Parameter Sharing

9 Feb 2018arXiv:1802.03268archive 2025-07-28

Hieu Pham, Melody Y. Guan, Barret Zoph, Quoc V. Le, Jeff Dean

We propose Efficient Neural Architecture Search (ENAS), a fast and inexpensive approach for automatic model design. In ENAS, a controller learns to discover neural network architectures by searching for an optimal subgraph within a large computational graph. The controller is trained with policy gradient to select a subgraph that maximizes the expected reward on the validation set. Meanwhile the model corresponding to the selected subgraph is trained to minimize a canonical cross entropy loss. Thanks to parameter sharing between child models, ENAS is fast: it delivers strong empirical performances using much fewer GPU-hours than all existing automatic model design approaches, and notably, 1000x less expensive than standard Neural Architecture Search. On the Penn Treebank dataset, ENAS discovers a novel architecture that achieves a test perplexity of 55.8, establishing a new state-of-the-art among all methods without post-training processing. On the CIFAR-10 dataset, ENAS designs novel architectures that achieve a test error of 2.89%, which is on par with NASNet (Zoph et al., 2018), whose test error is 2.65%.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1802.03268")

Code

Syntology Ran 18 of 38 code samples harvested from 13 repositories linked to this paper; 20 have no recorded run. Of those that ran: 1 ran · honoured contract; 2 ran · our draft was wrong; 4 ran · fixture could not drive it; 11 ran with no contract checked.

By repository: community (archive-listed): 38 samples from 13 repositories, 18 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

28 repositories listed; official and paper-mentioned ones first.

Ezereal/enas mentioned on GitHubtf report
HaiTMai/Time-Series-Forecast mentioned on GitHub report
MINGUKKANG/ENAS-Tensorflow mentioned on GitHubtf report
MINGUKKANG/PNU_Capstone_Design mentioned on GitHubtf report
MINGUKKANG/PNU_Termproject_ENAS mentioned on GitHubtf report
MengTianjian/enas-pytorch mentioned on GitHubpytorch report
RualPerez/AutoML mentioned on GitHubpytorch report
WillButAgain/ENAS mentioned on GitHubpytorch report
ahundt/enas mentioned on GitHubtf report
ahundt/renas mentioned on GitHubtf report
aymanshams07/enas_cifar10 mentioned on GitHubtf report
carpedm20/ENAS-pytorch mentioned on GitHubpytorch report
countif/enas_nni mentioned on GitHubtf report
cshannonn/blackscholes_nas mentioned on GitHub report
distrue/enas_tensorflow mentioned on GitHubtf report
f51980280/ENAS-Implement mentioned on GitHubpytorch report
google-research/google-research mentioned on GitHubtf report
guoyongcs/NAT mentioned on GitHubpytorch report
guoyongcs/NATv2 mentioned on GitHubpytorch report
invisibleForce/ENAS-Pytorch mentioned on GitHubpytorch report
melodyguan/enas mentioned on GitHubtf report
rutgerswiselab/autolossgen mentioned on GitHubpytorch report
zbyte64/pytorch-dagsearch mentioned on GitHubpytorch 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

38 samples harvested; 18 ran; 1 honoured the contract we drafted; 20 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
2ran · our draft was wrong
4ran · fixture could not drive it
11ran
20unverified

Licence: 25 of the 38 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 13 repositories linked to this paper, official or community; each sample names its own and says which. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

CNN MengTianjian/enas-pytorch/micro_child.py community (archive-listed) ran Apache-2.0 (permissive) · eab22bd84e3559c2 · report
Connector zbyte64/pytorch-dagsearch/dagsearch/dag.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · f4716de831df0059 · report
Controller google-research/google-research/enas_lm/src/controller.py community (archive-listed) ran Apache-2.0 (permissive) · c8730537c0104a93 · report
Graph zbyte64/pytorch-dagsearch/dagsearch/dag.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · acc4bf1454c8cc39 · report
HyperCell zbyte64/pytorch-dagsearch/dagsearch/dag.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · f356225b3b24f114 · report
Identity zbyte64/pytorch-dagsearch/dagsearch/dag.py community (archive-listed) ran fingerprinted no licence file found · pointer only · cefafcb412701ad5 · report
Interpolate zbyte64/pytorch-dagsearch/dagsearch/dag.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 2517ea4a652117c0 · report
LayerOp invisibleForce/ENAS-Pytorch/child_model.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · 83ec11e0af4a6100 · report
LossLayer rutgerswiselab/autolossgen/src/models/LossFormula.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · 53b29d9f18830ebd · report
NasLayer invisibleForce/ENAS-Pytorch/child_model.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · d733f480a9cc9d6a · report
PolicyNet RualPerez/AutoML/policy.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · 196af5ccc39415a7 · report
_lstm google-research/google-research/enas_lm/src/controller.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 0c0acd8344bfebad · report
_set_default_params google-research/google-research/enas_lm/src/controller.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · e2997adeaed89759 · report
count_model_params kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) ran · honoured contract no licence file found · pointer only · 4ad735e14ee6c0d4 · report
global_avg_pool kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) ran · fixture could not drive it no licence file found · pointer only · c122e163d2187416 · report
global_avgpool invisibleForce/ENAS-Pytorch/child_model.py community (archive-listed) ran · fixture could not drive it fingerprinted no licence file found · pointer only · 5bb25c6ad202549b · report
lstm ahundt/renas/enas/cifar10/general_controller.py community (archive-listed) ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 266925fcd482144b · report
stack_lstm ahundt/renas/enas/cifar10/general_controller.py community (archive-listed) ran · fixture could not drive it Apache-2.0 (permissive) · 19717b48f9cb0a9a · report
ChildModel invisibleForce/ENAS-Pytorch/child_model.py community (archive-listed) unverified no licence file found · pointer only · 864f33691502f6c2 · report
Controller ahundt/renas/enas/cifar10/general_controller.py community (archive-listed) unverified Apache-2.0 (permissive) · 6d91b9378fc9430c · report
GeneralController ahundt/renas/enas/cifar10/general_controller.py community (archive-listed) unverified Apache-2.0 (permissive) · ececc4762c60d0f9 · report
MicroChild kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 9031a7dc27f32e30 · report
MicroChild MINGUKKANG/ENAS-Tensorflow/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 1ce2168d55961e5d · report
MicroController MINGUKKANG/PNU_Termproject_ENAS/micro_controller.py community (archive-listed) unverified no licence file found · pointer only · 2b3cfe1fcdcd863b · report
MicroController ahundt/enas/enas/cifar10/micro_controller.py community (archive-listed) unverified Apache-2.0 (permissive) · 5bd590a434395d44 · report
MicroController distrue/enas_tensorflow/src/cifar10/micro_controller.py community (archive-listed) unverified Apache-2.0 (permissive) · 2c0648dbc9adf7c3 · report
MicroController melodyguan/enas/src/cifar10/micro_controller.py community (archive-listed) unverified Apache-2.0 (permissive) · 86474604afa3121b · report
Model kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 28a9ce542ac9b038 · report
Model MINGUKKANG/ENAS-Tensorflow/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 56d5391895065250 · report
View zbyte64/pytorch-dagsearch/dagsearch/dag.py community (archive-listed) unverified no licence file found · pointer only · d44cc1b50bd34d5c · report
_build_train_op google-research/google-research/enas_lm/src/controller.py community (archive-listed) unverified Apache-2.0 (permissive) · 848fb3ac5d7601b9 · report
batch_norm kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 5b9ee0a559cb458e · report
create_weight kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 6abbd474b4d1304b · report
create_weight MINGUKKANG/ENAS-Tensorflow/micro_child.py community (archive-listed) unverified no licence file found · pointer only · f2a563259f2890ae · report
drop_path kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 04864e18c26d7b4a · report
get_train_ops ahundt/renas/enas/cifar10/general_controller.py community (archive-listed) unverified Apache-2.0 (permissive) · c6454e6006e3888e · report
get_train_ops kaileymonn/Quantized-ENAS-ConvNets/micro_child.py community (archive-listed) unverified no licence file found · pointer only · 6176866b222836d8 · report
get_train_ops MINGUKKANG/ENAS-Tensorflow/micro_child.py community (archive-listed) unverified no licence file found · pointer only · dbdb4bb390fc9aaf · report

Tasks

Language ModellingNeural Architecture Search

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling Penn Treebank (Word Level) Efficient NAS Params 24M #31 of 43 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) Efficient NAS Test perplexity 58.6 #31 of 43 Archive leaderboard report
Language Modelling Penn Treebank (Word Level) Efficient NAS Validation perplexity 60.8 #31 of 43 Archive leaderboard report
Neural Architecture Search CIFAR-10 Image Classification ENAS + c/o Params 4.6M #17 of 19 Archive leaderboard report
Neural Architecture Search CIFAR-10 Image Classification ENAS + c/o Percentage error 2.89 #17 of 19 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

LSTMSigmoid ActivationSoftmaxTanh Activation

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