SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.12580 · 2023

Membership Inference Attacks against Synthetic Data through Overfitting Detection

arXiv · PDF · Open in the Atlas

Code that ran

4 of 6 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; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 2 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/2302.12580.json · badge: /api/ran/2302.12580.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
vanderschaarlab/domias canonical (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
FunctionStatusWhere it lives
GAN_leaks Ran
ran on a synthesized input
vanderschaarlab/domias/src/domias/baselines.py
code served (permissive licence) · get_code("883c31e7e7cb37f3")
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 ndarray, ndim 1, dtype family float, shape (2,) 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 numpy as np

# Create test data
# X_test: shape (2, 4) - 2 test samples of dimension 4
X_test = np.random.randn(2, 4).astype(np.float32)

# X_G: shape (3, 4) - 3 generated samples of dimension 4
X_G = np.random.randn(3, 4).astype(np.float32)

ARGS = {
    "X_test": X_test,
    "X_G": X_G,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) 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
compute_alpha_precision Ran
ran on a synthesized input
vanderschaarlab/domias/src/domias/metrics/evaluation.py
code served (permissive licence) · get_code("97176f9c559cc45f")
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 numpy as np

torch.manual_seed(0)

# Create small, distinct-sized synthetic data
# real_data: shape (2, 4) - 2 samples, 4 dimensions
real_data = torch.randn(2, 4).numpy()

# synthetic_data: shape (3, 4) - 3 samples, 4 dimensions (distinct from real_data size)
synthetic_data = torch.randn(3, 4).numpy()

# emb_center: shape (4,) - center point in embedding space
emb_center = torch.randn(4).numpy()

# alternative_coverage: boolean flag
alternative_coverage = False

ARGS = {
    'real_data': real_data,
    'synthetic_data': synthetic_data,
    'emb_center': emb_center,
    'alternative_coverage': alternative_coverage,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, 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
d Ran
ran on a synthesized input
vanderschaarlab/domias/src/domias/baselines.py
code served (permissive licence) · get_code("8f9c330f5ce6bd85")
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 ndarray, ndim 1, dtype family float, shape (3,) 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 numpy as np

# The function d computes squared Euclidean distances.
# It handles two cases:
# 1. If X is 1D, it computes distance between X and each row of Y
# 2. If X is 2D, it computes pairwise distances between rows of X and rows of Y

# For case 1: X is 1D (shape: (4,)), Y is 2D (shape: (3, 4))
# Result should be 1D array of shape (3,)

# For case 2: X is 2D (shape: (2, 4)), Y is 2D (shape: (3, 4))
# Result should be 2D array of shape: (2, 3)

# Create test data for the 1D case
X_1d = np.array([1.0, 2.0, 3.0, 4.0])
Y = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [2.0, 3.0, 4.0, 5.0],
    [0.0, 0.0, 0.0, 0.0]
])

ARGS = {
    "X": X_1d,
    "Y": Y
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (3,) 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
load_dataset Ran
ran on a synthesized input
vanderschaarlab/domias/src/domias/bnaf/density_estimation.py
code served (permissive licence) · get_code("7bf6fc70b69f9c4b")
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, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type and ndim 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 numpy as np

torch.manual_seed(0)

# Create small numpy arrays for training, validation, and test data
# Using distinct shapes to make dimensions clear: (2, 4) for samples x features
data_train = np.random.randn(2, 4).astype(np.float32)
data_valid = np.random.randn(2, 4).astype(np.float32)
data_test = np.random.randn(2, 4).astype(np.float32)

# Set device to CPU only
device = torch.device('cpu')

# Batch dimension
batch_dim = 2

ARGS = {
    'data_train': data_train,
    'data_valid': data_valid,
    'data_test': data_test,
    'device': device,
    'batch_dim': batch_dim
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
d_min Not yet run
unverified (harvested, no recorded run)
vanderschaarlab/domias/src/domias/baselines.py
code served (permissive licence) · get_code("b2b257639c00245e")
How it ran

no invocation recorded.

save_model Not yet run
unverified (harvested, no recorded run)
vanderschaarlab/domias/src/domias/bnaf/density_estimation.py
code served (permissive licence) · get_code("98ed3d40f8a5dfa5")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 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.

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

Data is the foundation of most science. Unfortunately, sharing data can be obstructed by the risk of violating data privacy, impeding research in fields like healthcare. Synthetic data is a potential solution. It aims to generate data that has the same distribution as the original data, but that does not disclose information about individuals. Membership Inference Attacks (MIAs) are a common privacy attack, in which the attacker attempts to determine whether a particular real sample was used for training of the model. Previous works that propose MIAs against generative models either display low performance -- giving the false impression that data is highly private -- or need to assume access to internal generative model parameters -- a relatively low-risk scenario, as the data publisher often only releases synthetic data, not the model. In this work we argue for a realistic MIA setting that assumes the attacker has some knowledge of the underlying data distribution. We propose DOMIAS, a density-based MIA model that aims to infer membership by targeting local overfitting of the generative model. Experimentally we show that DOMIAS is significantly more successful at MIA than previous work, especially at attacking uncommon samples. The latter is disconcerting since these samples may correspond to underrepresented groups. We also demonstrate how DOMIAS' MIA performance score provides an interpretable metric for privacy, giving data publishers a new tool for achieving the desired privacy-utility trade-off in their synthetic data.

For agents

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

get_harvested_code_for_paper("2302.12580")
get_code_for_paper("2302.12580")
have("2302.12580")

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

curl https://syntology.ai/api/ran/2302.12580.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.12580.svg)](https://syntology.ai/paper/2302.12580)

Connect an agent — have() is free.