Papers › Deep Architecture Connectivity Matters for Its Convergence: A Fine-Grained Analysis

Deep Architecture Connectivity Matters for Its Convergence: A Fine-Grained Analysis

11 May 2022arXiv:2205.05662archive 2025-07-28

Wuyang Chen, Wei Huang, Xinyu Gong, Boris Hanin, Zhangyang Wang

Advanced deep neural networks (DNNs), designed by either human or AutoML algorithms, are growing increasingly complex. Diverse operations are connected by complicated connectivity patterns, e.g., various types of skip connections. Those topological compositions are empirically effective and observed to smooth the loss landscape and facilitate the gradient flow in general. However, it remains elusive to derive any principled understanding of their effects on the DNN capacity or trainability, and to understand why or in which aspect one specific connectivity pattern is better than another. In this work, we theoretically characterize the impact of connectivity patterns on the convergence of DNNs under gradient descent training in fine granularity. By analyzing a wide network's Neural Network Gaussian Process (NNGP), we are able to depict how the spectrum of an NNGP kernel propagates through a particular connectivity pattern, and how that affects the bound of convergence rates. As one practical implication of our results, we show that by a simple filtration on "unpromising" connectivity patterns, we can trim down the number of models to evaluate, and significantly accelerate the large-scale neural architecture search without any overhead. Code is available at: https://github.com/VITA-Group/architecture_convergence.

PaperPDFCodeCode Syntology ran

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 21 code samples harvested from 2 repositories linked to this paper: 19 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 2 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 21 samples from 2 repositories, 19 ran (16 with no instrument failure; 3 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.

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. 4 of the 19 samples that ran have no recorded date.

chenwydj/architecture_convergence officialmentioned in papermentioned on GitHubpytorchMITSyntology: 20 samples, 18 ran (2 instrument-failed) · 2 unverified report
vita-group/architecture_convergence officialmentioned in papermentioned on GitHubpytorchMITSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 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.

Code Syntology ran Syntology

21 samples harvested: 19 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 2 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 2 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.

2ran · violated contract
3ran · our draft was wrong
14ran
2unverified

Licence: 2 of the 21 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, 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 2 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.

chenwydj/architecture_convergence official repository 20 samples, 18 ran (2 instrument-failed) · 2 unverified
calculate_md5 in 16 papers nas-without-training/datasets/DownsampledImageNet.py ran · our draft was wrong MIT (permissive) · ad0636a128295d5a · report fixture: signature fpath:str/0/str, chunk_size:int/0/int · returned str · level recorded, date not recorded
check_md5 in 15 papers nas-without-training/datasets/DownsampledImageNet.py ran · violated contract MIT (permissive) · 380a00b777aad655 · report fixture: signature fpath:str/0/str, md5:str/0/str · returned bool · level recorded, date not recorded
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
mlp_code/utils.py ran MIT (permissive) · fce4d8316ff3d185 · report fixture: signature output:Tensor/2/float, label:Tensor/1/int, num_class:int/0/int, topk:tuple/1/int · returned tuple · level recorded 2026-09-24
check_integrity in 37 papers nas-without-training/datasets/DownsampledImageNet.py ran · violated contract fingerprinted MIT (permissive) · e947a6effcaaf32b · report fixture: signature fpath:str/0/str, md5:str/0/str · returned bool · level recorded, date not recorded
edge2value nas-without-training/collect_arch_data.py ran fingerprinted MIT (permissive) · d9b7f522f765561e · report fixture: signature op_str:str/0/str · returned int · level recorded 2026-09-24
find_all_paths nas-without-training/collect_arch_data.py ran MIT (permissive) · c2c1522bfc9c4a8b · report fixture: signature Aff:ndarray/2/float, all_paths:list/1/empty, curr_path:list/1/empty, curr_pos:int/0/int, end_pos:int/0/int · returned list · level recorded 2026-09-24
find_all_paths_3d_prob TENAS/prune_tenas.py ran MIT (permissive) · 5cc4e76addc371ea · report fixture: signature Aff:ndarray/3/float, all_paths:list/1/empty, curr_path:list/1/empty, curr_prob:float/0/float, curr_pos:int/0/int, end_pos:int/0/int · returned list · level recorded 2026-09-24
format_time in 140 papers mlp_code/utils.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · 8d7ec010c29e813b · report fixture: signature seconds:float/0/float · returned str · level recorded, date not recorded
get_batch_jacobian nas-without-training/score_networks.py ran MIT (permissive) · e7c03b50082d9c61 · report fixture: signature net:callable/0/?, x:Tensor/2/float, target:Tensor/2/float, device:str/0/str, args:NoneType/0/? · returned tuple · level recorded 2026-09-24
get_mean_and_std in 55 papers mlp_code/utils.py ran · metamorphic tier: well formed MIT (permissive) here; the stored copy is pointer only · 3cbdd3b190b16350 · report fixture: signature dataset:MinimalDataset/0/? · returned tuple · level recorded 2026-09-22
get_score_func nas-without-training/scores.py ran MIT (permissive) · 16139ec12e4595a3 · report fixture: signature score_name:str/0/str · returned function · level recorded 2026-09-24
hooklogdet nas-without-training/scores.py ran MIT (permissive) · 8d9701991ac208d4 · report fixture: signature K:ndarray/2/float, labels:ndarray/1/int · returned float64 · level recorded 2026-09-26
init_model in 38 papers TENAS/prune_tenas.py ran MIT (permissive) · e1591bd29b12e21d · report fixture: signature model:callable/0/?, method:str/0/str · returned SmallModel · level recorded 2026-09-24
make_colours_by_hue nas-without-training/plot_scores.py ran MIT (permissive) · 2cd6cda6c652de6d · report fixture: signature h:float/0/float, v:float/0/float · returned list · level recorded 2026-09-24
prepare_logger mlp_code/logger.py ran MIT (permissive) · a87bbc139870cdc0 · report fixture: signature xargs:MockArgs/0/? · returned Logger · level recorded 2026-09-24
random_score nas-without-training/scores.py ran MIT (permissive) · 3aa2047d05e1a518 · report fixture: signature jacob:ndarray/2/float, label:ndarray/1/int · returned float · level recorded 2026-09-24
round_to TENAS/prune_tenas.py ran MIT (permissive) · 65d81629d7179cf5 · report fixture: signature number:float/0/float, precision:int/0/int, eps:float/0/float · returned float · level recorded 2026-09-24
svhn_dataloaders mlp_code/dataset.py ran MIT (permissive) · cc11bb618b200b77 · report fixture: returned tuple · level recorded 2026-09-24
get_cifar_test_loader mlp_code/dataset.py unverified MIT (permissive) · 44f160db0172cb70 · report
get_svhn_test_loader mlp_code/dataset.py unverified MIT (permissive) · aa037b9bfa339105 · report
vita-group/architecture_convergence official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
get_flop_loss nas-without-training/autodl/procedures/search_main.py ran · our draft was wrong fingerprinted MIT (permissive) · 5aa7497aa00af169 · report fixture: signature expected_flop:Tensor/1/float, flop_cur:float/0/float, flop_need:float/0/float, flop_tolerant:float/0/float · returned tuple · level recorded 2026-09-22

Tasks

AutoMLNeural Architecture Search

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Gaussian Process

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