Papers › Surprisingly Strong Performance Prediction with Neural Graph Features

Surprisingly Strong Performance Prediction with Neural Graph Features

25 Apr 2024arXiv:2404.16551archive 2025-07-28

Gabriela Kadlecová, Jovita Lukasik, Martin Pilát, Petra Vidnerová, Mahmoud Safari, Roman Neruda, Frank Hutter

Performance prediction has been a key part of the neural architecture search (NAS) process, allowing to speed up NAS algorithms by avoiding resource-consuming network training. Although many performance predictors correlate well with ground truth performance, they require training data in the form of trained networks. Recently, zero-cost proxies have been proposed as an efficient method to estimate network performance without any training. However, they are still poorly understood, exhibit biases with network properties, and their performance is limited. Inspired by the drawbacks of zero-cost proxies, we propose neural graph features (GRAF), simple to compute properties of architectural graphs. GRAF offers fast and interpretable performance prediction while outperforming zero-cost proxies and other common encodings. In combination with other zero-cost proxies, GRAF outperforms most existing performance predictors at a fraction of the cost.

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 22 code samples harvested from 1 repository linked to this paper: 19 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 0 violated, 8 with no contract checked; 8 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 8 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 22 samples from 1 repository, 19 ran (11 with no instrument failure; 8 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 on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

gabikadlecova/zc_combine officialmentioned in paperMITSyntology: 22 samples, 19 ran (8 instrument-failed) · 3 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

22 samples harvested: 19 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 0 violated, 8 with no contract checked; 8 where Syntology's instrument failed) · 3 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.

3ran · honoured contract
8ran · our draft was wrong
8ran
3unverified

Licence: 0 of the 22 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 gabikadlecova/zc_combine. “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.

gabikadlecova/zc_combine official repository 22 samples, 19 ran (8 instrument-failed) · 3 unverified
aggregate_isomorphic_scores zc_combine/fixes/utils.py ran MIT (permissive) · a0cd5cb2d67c483f · report fixture: signature data:DataFrame/0/?, proxy_cols:list/1/str, agg_func:callable/0/?, val_accs_std:int/0/int, score_std:int/0/int, net_key:str/0/str · returned DataFrame · level recorded 2026-09-24
apply_layer_dropout_relu zc_combine/predictors/gnn.py ran MIT (permissive) · 8f09f06ced25bf21 · report fixture: signature x:Tensor/2/float, layers:callable/0/?, p:float/0/float, training:bool/0/bool · returned Tensor · level recorded 2026-09-24
compute_feature_df zc_combine/features/dataset.py ran MIT (permissive) · 076a7b43b581f744 · report fixture: signature nets:dict/0/?, feature_func:callable/0/?, name:str/0/str · returned DataFrame · level recorded 2026-09-24
count_edges zc_combine/features/nasbench101.py ran · honoured contract MIT (permissive) · feee61f724f4fe31 · report fixture: signature net:list/1/NoneType · returned int · level recorded 2026-09-24
count_ops zc_combine/features/base.py ran · our draft was wrong MIT (permissive) · fc531b68e3daae1c · report fixture: signature net:tuple/1/dict, to_names:bool/0/bool · returned dict · level recorded 2026-09-24
count_ops_opnodes zc_combine/features/base.py ran · our draft was wrong MIT (permissive) · 820f496e1ce6d0c5 · report fixture: signature net:tuple/1/list, to_names:bool/0/bool · returned dict · level recorded 2026-09-24
count_switches zc_combine/features/tnb101_macro.py ran · our draft was wrong MIT (permissive) · 4074afd5d8f8712e · report fixture: signature net:list/1/dict · returned dict · level recorded 2026-09-24
edge zc_combine/fixes/utils.py ran MIT (permissive) · da5ba5a4db691671 · report fixture: signature edge_map:Tensor/2/int, op_list:list/1/str, i:int/0/int, j:int/0/int · returned str · level recorded 2026-09-24
filter_by_range zc_combine/fixes/operations.py ran MIT (permissive) · 72ce8a466cc40e81 · report fixture: signature df:DataFrame/0/?, zc:str/0/str, min:float/0/float, max:float/0/float · returned DataFrame · level recorded 2026-09-24
get_state_at_i zc_combine/features/tnb101_macro.py ran · our draft was wrong MIT (permissive) · 38fba84176f9db88 · report fixture: signature net:list/1/dict · returned dict · level recorded 2026-09-24
load_from_config zc_combine/features/dataset.py ran MIT (permissive) · 9a4c1d594afc3573 · report fixture: signature nets:dict/0/?, func_cfg:dict/0/?, func_dict:dict/0/? · returned DataFrame · level recorded 2026-09-24
log_dataset_args zc_combine/utils/args_utils.py ran · our draft was wrong MIT (permissive) · 83112c2bbada9b2c · report fixture: signature args:dict/0/? · returned dict · level recorded 2026-09-24
op_on_pos zc_combine/features/nasbench201.py ran · our draft was wrong MIT (permissive) · 19f379c601ea9e9f · report fixture: signature net:tuple/1/dict · returned dict · level recorded 2026-09-24
parse_and_read_args zc_combine/utils/args_utils.py ran · our draft was wrong MIT (permissive) · 7d6da07092b48621 · report fixture: signature parser:ArgumentParser/0/? · returned dict · level recorded 2026-09-24
parse_ops_nb201 zc_combine/fixes/operations.py ran · honoured contract MIT (permissive) · 963aae00115ab847 · report fixture: signature df:DataFrame/0/?, net_key:str/0/str · returned list · level recorded 2026-09-24
parse_ops_nb301 zc_combine/fixes/operations.py ran · honoured contract MIT (permissive) · c79827e617a35632 · report fixture: signature df:DataFrame/0/? · returned list · level recorded 2026-09-24
remap_values zc_combine/features/conversions.py ran fingerprinted MIT (permissive) · da732986788c2dc3 · report fixture: signature c:int/0/int, val1:int/0/int, val2:int/0/int · returned int · level recorded 2026-09-24
to_graph zc_combine/features/base.py ran · our draft was wrong MIT (permissive) · 73cae9b8d56e5f84 · report fixture: signature edges:list/1/tuple · returned DiGraph · level recorded 2026-09-24
zero_outgoing zc_combine/fixes/utils.py ran MIT (permissive) · fd21b29f71363669 · report fixture: signature net:list/1/int, edge_map:Tensor/2/int, zero_op:int/0/int · returned list · level recorded 2026-09-24
get_feature_dataset zc_combine/features/dataset.py unverified MIT (permissive) · 6a2d7c76fc5717cd · report
to_dataset zc_combine/predictors/gnn.py unverified MIT (permissive) · 2c6ead1f8f01739f · report
train in 522 papers zc_combine/predictors/gnn.py unverified MIT (permissive) · 21314b00e73896ce · report

Tasks

Neural Architecture SearchPrediction

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SPEED

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