SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.02191 · NeurIPS · 2024

Attack-Aware Noise Calibration for Differential Privacy

Georgios Kaissis, Bogdan Kulynych, Juan Gomez, Carmela Troncoso

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 1 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/2407.02191.json · badge: /api/ran/2407.02191.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
bogdan-kulynych/riskcal canonical (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
FunctionStatusWhere it lives
get_fixed_point_case1 Ran
ran, honoured the contract we drafted
bogdan-kulynych/riskcal/src/riskcal/analysis/rdp/pure.py
code served (permissive licence) · get_code("46b87ad43702e7e2")
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 float, ndim 0, dtype family float, shape () 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
import numpy as np

# Analyzing the function get_fixed_point_case1:
# Parameters:
# - order: int, used in mathematical expressions (order > 1 assumed)
# - upper: float, used as threshold in comparison
# - tol: float, tolerance for bisection (default 1e-07)
#
# The function uses these parameters in:
# - F_mid = (1 - mid) ** (1 - order) * mid ** order + mid ** (1 - order) * (1 - mid) ** order
# - if F_mid > upper: comparison
# - while high - low > tol: tolerance check
#
# All parameters are scalars (int or float), so straightforward to construct.

ARGS = {
    "order": 2,      # order > 1 as required by docstring
    "upper": 0.5,    # some threshold value between 0 and 1 for the fixed point
    "tol": 1e-07     # default tolerance
}
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 float, ndim 0, dtype family float, shape () — shape () 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
get_fixed_point_case2 Ran
ran, honoured the contract we drafted
bogdan-kulynych/riskcal/src/riskcal/analysis/rdp/pure.py
code served (permissive licence) · get_code("324a3e29f2a25b00")
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 float, ndim 0, dtype family float, shape () 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.

from math import fabs, pow, exp, log, log1p

# ARGS for get_fixed_point_case2
ARGS = {
    "epsilon": 0.5,
    "tol": 1e-07
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () 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
inverse_monotone_function Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bogdan-kulynych/riskcal/src/riskcal/utils.py
code served (permissive licence) · get_code("0e6110126ad4d01f")
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 float, ndim 0, dtype family float, shape () 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 math

# Define a simple monotonic function to invert
# For a decreasing function: f(x) = 10 - x^2
def decreasing_func(x: float) -> float:
    return 10.0 - x**2

# For an increasing function: f(x) = x^2
def increasing_func(x: float) -> float:
    return x**2

# Test with decreasing function
# We want to find x such that f(x) ≈ 6
# For f(x) = 10 - x^2 = 6, x^2 = 4, so x = 2
ARGS = {
    "f": decreasing_func,
    "f_target": 6.0,
    "bounds": (0.0, 4.0),
    "func_threshold": 0.001,
    "param_threshold": 0.001,
    "max_iter": 100,
    "increasing": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () 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
get_FNR_case1_vec Not yet run
unverified (harvested, no recorded run)
bogdan-kulynych/riskcal/src/riskcal/analysis/rdp/pure.py
code served (permissive licence) · get_code("8352363ee57606c1")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 4 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 4 later) and 0 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
get_FNR_case1_vec 1 other paper
1 later
2507.06969 (2025-07)
Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy
get_fixed_point_case1 1 other paper
1 later
2507.06969 (2025-07)
Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy
get_fixed_point_case2 1 other paper
1 later
2507.06969 (2025-07)
Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy
inverse_monotone_function 1 other paper
1 later
2507.06969 (2025-07)
Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy

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

Differential privacy (DP) is a widely used approach for mitigating privacy risks when training machine learning models on sensitive data. DP mechanisms add noise during training to limit the risk of information leakage. The scale of the added noise is critical, as it determines the trade-off between privacy and utility. The standard practice is to select the noise scale to satisfy a given privacy budget ε. This privacy budget is in turn interpreted in terms of operational attack risks, such as accuracy, sensitivity, and specificity of inference attacks aimed to recover information about the training data records. We show that first calibrating the noise scale to a privacy budget ε, and then translating ε to attack risk leads to overly conservative risk assessments and unnecessarily low utility. Instead, we propose methods to directly calibrate the noise scale to a desired attack risk level, bypassing the step of choosing ε. For a given notion of attack risk, our approach significantly decreases noise scale, leading to increased utility at the same level of privacy. We empirically demonstrate that calibrating noise to attack sensitivity/specificity, rather than ε, when training privacy-preserving ML models substantially improves model accuracy for the same risk level. Our work provides a principled and practical way to improve the utility of privacy-preserving ML without compromising on privacy.

For agents

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

get_harvested_code_for_paper("2407.02191")
get_code_for_paper("2407.02191")
have("2407.02191")

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

curl https://syntology.ai/api/ran/2407.02191.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.02191.svg)](https://syntology.ai/paper/2407.02191)

Connect an agent — have() is free.