SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.07954 · ICML · 2021

AlphaNet: Improved Training of Supernets with Alpha-Divergence

Chengyue Gong, Qiang Liu, Meng Li, Vikas Chandra, Dilin Wang

arXiv · PDF · Open in the Atlas

Code that ran

16 of 31 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 4 failures of our instrument; 15 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2102.07954.json · badge: /api/ran/2102.07954.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
facebookresearch/AlphaNet canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
facebookresearch/AttentiveNAS named in the paper
Papers with Code archive
14 ran · 15 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
ConvBnActLayer Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("a7e70ce25e3ac2a4")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

DynamicBatchNorm2d Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("a0efb5eabd43bfb3")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

DynamicLinear Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("86a56f5a600b5419")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

DynamicLinearLayer Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("94f0bda6ed5a2e9d")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

DynamicPointConv2d Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("df25f4eefb96b548")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

IdentityLayer Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("0257146dc3978be2")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

LinearLayer Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("c3abc01ffd8662f7")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

ShortcutLayer Ran
ran on a synthesized input
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("1c4df2c55cd70330")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

build_activation Ran
ran; our contract draft was wrong, not the code
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("6150fe6395eb4ee3")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

drop_connect Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("6053cb1369b46c73")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

f_divergence Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
facebookresearch/AlphaNet/loss_ops.py
pointer only (licence: NOASSERTION) · get_code("bd606208d8003d1e")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AlphaNet (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_net_device Ran
ran; our contract draft was wrong, not the code
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("0e1ee8a546f09486")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_same_padding Ran
ran, honoured the contract we drafted
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("71cbea6f98bde1f9")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

int2list Ran
ran, honoured the contract we drafted
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("35ee83149f0affc7")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

make_divisible Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from JiahuiYu/slimmable_networks
pointer only; the stored copy get_code serves, from JiahuiYu/slimmable_networks, is pointer only (licence: NOASSERTION) · get_code("3356d3566f053dcc")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

sub_filter_start_end Ran
ran, honoured the contract we drafted
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("00fdaefcd3c7d435")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

AllReduce Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("0b4bd61096ddb694")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

AttentiveNasDynamicModel Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("06d08882ecf1c68a")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

AttentiveNasStaticModel Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("8ff71dfd928cae27")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

DynamicConvBnActLayer Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("39b14fb632110a3f")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

DynamicMBConvLayer Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("c582a9fdd53c2f23")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

DynamicSE Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("4c386a2f1dbb59bd")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

DynamicSeparableConv2d Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("7949f6e86881267d")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

DynamicShortcutLayer Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("bb02270a794b549e")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

MBInvertedConvLayer Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("1305fda939e9811f")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

MobileInvertedResidualBlock Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("d7620e552fd063af")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

MyModule Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("6669bff300b9aa8f")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

MyNetwork Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("f6b0ce7520ccf192")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

SELayer Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("a8a91d970372f10e")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

copy_bn Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("551a6373b2274b70")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

set_layer_from_config Not yet run
unverified (harvested, no recorded run)
facebookresearch/AttentiveNAS/models/attentive_nas_dynamic_model.py
pointer only (licence: NOASSERTION) · get_code("102325b4dff7f585")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, facebookresearch/AttentiveNAS (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 31 samples lifted from this paper's linked repositories, 31 carry a function identity and 0 do not. Of the 31 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 29 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
make_divisible
brick of this paper's assembly: helper
4 other papers
3 dated earlier, 1 later
1812.08928 (2018-12)
Slimmable Neural Networks
1903.05134 (2019-03)
Universally Slimmable Networks and Improved Training Techniques
2012.03153 (2020-12)
Any-Width Networks
and 1 more not named here
get_same_padding
brick of this paper's assembly: helper
1 other paper
1 dated earlier
1908.09791 (2019-08)
Once-for-All: Train One Network and Specialize it for Efficient Deployment

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Weight-sharing neural architecture search (NAS) is an effective technique for automating efficient neural architecture design. Weight-sharing NAS builds a supernet that assembles all the architectures as its sub-networks and jointly trains the supernet with the sub-networks. The success of weight-sharing NAS heavily relies on distilling the knowledge of the supernet to the subnetworks. However, we find that the widely used distillation divergence, i.e., KL divergence, may lead to student sub-networks that overestimate or under-estimate the uncertainty of the teacher supernet, leading to inferior performance of the sub-networks. In this work, we propose to improve the supernet training with a more generalized α-divergence. By adaptively selecting the α-divergence, we simultaneously prevent the over-estimation or under-estimation of the uncertainty of the teacher model. We apply the proposed α-divergence based supernets training to both slimmable neural networks and weight-sharing NAS, and demonstrate significant improvements. Specifically, our discovered model family, AlphaNet, outperforms prior-art models on a wide range of FLOPs regimes, including BigNAS, Once-for-All networks, and AttentiveNAS. We achieve ImageNet top-1 accuracy of 80.0% with only 444M FLOPs. Our code and pretrained models are available at https://github.com/ facebookresearch/AlphaNet.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2102.07954")
get_code_for_paper("2102.07954")
have("2102.07954")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2102.07954.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2102.07954.svg)](https://syntology.ai/paper/2102.07954)

Connect an agent — have() is free.