SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2211.06302 · NeurIPS · 2022

GCondNet: A Novel Method for Improving Neural Networks on Small High-Dimensional Tabular Data

Pietro Liò, Nikola Simidjievski, Andrei Margeloiu, Mateja Jamnik

arXiv · PDF · Open in the Atlas

Code that ran

13 of 20 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (5 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 4 failures of our instrument; 7 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/2211.06302.json · badge: /api/ran/2211.06302.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
andreimargeloiu/gcondnet canonical (classifier label)
classifier basis: classifier_2025
6 ran · 1 no recorded run
lasso-net/lassonet unrelated (classifier label)
classifier basis: classifier_2025
3 ran · 3 no recorded run
andreimargeloiu/WPFS community
Papers with Code archive
3 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 1 no recorded run
jjfeng/spinn unrelated (classifier label)
classifier basis: classifier_2025
0 ran · 2 no recorded run
FunctionStatusWhere it lives
GNNWeightPredictorNetwork Ran
ran on a synthesized input · constructed an object; no result was computed
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("42ddf24702948f06")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

GeneralNeuralNetwork Ran
ran on a synthesized input · constructed an object; no result was computed
andreimargeloiu/WPFS/src/models.py
pointer only (licence: NONE) · get_code("87ab77b44f33747f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/WPFS when it was checked. 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.

SparsityNetwork Ran
ran on a synthesized input · constructed an object; no result was computed
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("adfcee37607f04fd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

WeightPredictorNetwork Ran
ran on a synthesized input · constructed an object; no result was computed
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("6ba9ae7b6b50cd7e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

ZeroWeightPredictorNetwork Ran
ran on a synthesized input · constructed an object; no result was computed
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("43e63844eb5d2d93")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

compute_graph_embeddings Ran
ran, honoured the contract we drafted
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("6ff4ab3a68a7a9b5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

create_linear_layers Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
andreimargeloiu/WPFS/src/models.py
pointer only (licence: NONE) · get_code("871a4aab89f8e63a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/WPFS when it was checked. 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_labels_lists Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from andreimargeloiu/wpfs
pointer only; the stored copy get_code serves, from andreimargeloiu/wpfs, is pointer only (licence: NONE) · get_code("cda491bdda02d633")
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.

linear_interpolation_coefficient Ran
ran, honoured the contract we drafted
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("8b1858551f6b22c4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

prox Ran
ran; our contract draft was wrong, not the code
lasso-net/lassonet/lassonet/model.py
code served (permissive licence) · get_code("375e015ef10a230a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Analyze the function signature and body:
# - v: shape (m,) or (m, batches) - used with unsqueeze, norm, gather
# - u: shape (k,) or (k, batches) - used with unsqueeze, abs, sort, sign
# - lambda_: scalar - used in arithmetic
# - lambda_bar: scalar - used in soft_threshold
# - M: scalar - used in arithmetic

# Create small test tensors with distinct dimensions
m = 5  # dimension for v
k = 4  # dimension for u
batches = 2  # batch dimension

v = torch.randn(m)  # shape (m,)
u = torch.randn(k)  # shape (k,)
lambda_ = 0.5  # scalar regularization parameter
lambda_bar = 0.3  # scalar threshold parameter
M = 0.1  # scalar matrix parameter

ARGS = {
    "v": v,
    "u": u,
    "lambda_": lambda_,
    "lambda_bar": lambda_bar,
    "M": M,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
reshape_batch Ran
ran; our contract draft was wrong, not the code
andreimargeloiu/WPFS/src/models.py
pointer only (licence: NONE) · get_code("ff587bdefe78c65d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/WPFS when it was checked. 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.

sign_binary Ran
ran; our contract draft was wrong, not the code
lasso-net/lassonet/lassonet/model.py
code served (permissive licence) · get_code("72eb093eb143edd8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function sign_binary takes a single parameter x
# It expects x to be a torch tensor since it uses torch operations on it
# Creating a small tensor with distinct dimensions (2 x 4)
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
soft_threshold Ran
ran, honoured the contract we drafted
lasso-net/lassonet/lassonet/model.py
code served (permissive licence) · get_code("3f9208771cec0b61")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# soft_threshold takes two parameters:
# l: a threshold value (scalar)
# x: a tensor to apply soft thresholding to

l = 0.5
x = torch.randn(2, 4)

ARGS = {
    "l": l,
    "x": x,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
FirstLinearLayer Not yet run
unverified (harvested, no recorded run)
andreimargeloiu/gcondnet/src/models.py
pointer only (licence: NONE) · get_code("9343413040a43530")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, andreimargeloiu/gcondnet when it was checked. 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.

LassoNet Not yet run
unverified (harvested, no recorded run)
lasso-net/lassonet/lassonet/model.py
code served (permissive licence) · get_code("a8a3e7989cfd2596")
How it ran

no invocation recorded.

NeuralNetwork Not yet run
unverified (harvested, no recorded run)
jjfeng/spinn/neural_network.py
pointer only (licence: NONE) · get_code("0e259c599f65cde2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jjfeng/spinn when it was checked. 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.

NeuralNetworkParams Not yet run
unverified (harvested, no recorded run)
jjfeng/spinn/neural_network.py
pointer only (licence: NONE) · get_code("69d8f664512ed227")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jjfeng/spinn when it was checked. 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.

detach_tensors Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from andreimargeloiu/wpfs
pointer only; the stored copy get_code serves, from andreimargeloiu/wpfs, is pointer only (licence: NONE) · get_code("ffdfda6bfc1ce5a8")
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. No invocation recorded.

inplace_group_prox Not yet run
unverified (harvested, no recorded run)
lasso-net/lassonet/lassonet/model.py
code served (permissive licence) · get_code("dbb94584aa252f58")
How it ran

no invocation recorded.

inplace_prox Not yet run
unverified (harvested, no recorded run)
lasso-net/lassonet/lassonet/model.py
code served (permissive licence) · get_code("6c397ca47373e207")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 5 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 2 later) and 15 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
detach_tensors
brick of this paper's assembly: helper
2 other papers
1 the same month, 1 later
2211.15616 (2022-11)
Weight Predictor Network with Feature Selection for Small Sample Tabular Biomedical Data
2306.12330 (2023-06)
ProtoGate: Prototype-based Neural Networks with Global-to-local Feature Selection for Tabular Biomedical Data
get_labels_lists
brick of this paper's assembly: helper
2 other papers
1 the same month, 1 later
2211.15616 (2022-11)
Weight Predictor Network with Feature Selection for Small Sample Tabular Biomedical Data
2306.12330 (2023-06)
ProtoGate: Prototype-based Neural Networks with Global-to-local Feature Selection for Tabular Biomedical Data
prox 1 other paper
1 dated earlier
1907.12207 (2019-07)
LassoNet: A Neural Network with Feature Sparsity
sign_binary 1 other paper
1 dated earlier
1907.12207 (2019-07)
LassoNet: A Neural Network with Feature Sparsity
soft_threshold 1 other paper
1 dated earlier
1907.12207 (2019-07)
LassoNet: A Neural Network with Feature Sparsity

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

Neural networks often struggle with high-dimensional but small sample-size tabular datasets. One reason is that current weight initialisation methods assume independence between weights, which can be problematic when there are insufficient samples to estimate the model's parameters accurately. In such small data scenarios, leveraging additional structures can improve the model's performance and training stability. To address this, we propose GCondNet, a general approach to enhance neural networks by leveraging implicit structures present in tabular data. We create a graph between samples for each data dimension, and utilise Graph Neural Networks (GNNs) to extract this implicit structure, and for conditioning the parameters of the first layer of an underlying predictor network. By creating many small graphs, GCondNet exploits the data's high-dimensionality, and thus improves the performance of an underlying predictor network. We demonstrate GCondNet's effectiveness on 12 real-world datasets, where it outperforms 14 standard and state-of-the-art methods. The results show that GCondNet is a versatile framework for injecting graph-regularisation into various types of neural networks, including MLPs and tabular Transformers. Code is available at https://github.com/andreimargeloiu/GCondNet.

For agents

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

get_harvested_code_for_paper("2211.06302")
get_code_for_paper("2211.06302")
have("2211.06302")

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

curl https://syntology.ai/api/ran/2211.06302.json

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

[![Syntology run record](https://syntology.ai/api/ran/2211.06302.svg)](https://syntology.ai/paper/2211.06302)

Connect an agent — have() is free.