SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.14838 · 2021

Evaluating Privacy-Preserving Machine Learning in Critical Infrastructures: A Case Study on Time-Series Classification

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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 7 have no contract verdict; 0 failures 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/2111.14838.json · badge: /api/ran/2111.14838.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
DominiqueMercier/PPML-TSA canonical (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
FunctionStatusWhere it lives
build_mean_dict Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/mean_cr_utils.py
code served (permissive licence) · get_code("b11acd37a1a19ede")
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 dict. 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)

# Analyze the function to understand what arguments it needs:
# - structure: iterable of items, each is either a str or a tuple/list starting with str
# - means: iterable of numeric values
# - stds: iterable of numeric values
# All three should have the same length and be zippable

structure = [
    'metric1',
    'metric2',
    ('group1', 'submetric1'),
    ('group1', 'submetric2'),
    ('group2', 'submetric3')
]

means = [0.8542, 0.7231, 0.9123, 0.6789, 0.5432]

stds = [0.0234, 0.0456, 0.0123, 0.0567, 0.0345]

ARGS = {
    'structure': structure,
    'means': means,
    'stds': stds
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
build_mean_dict Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/AL/mean_cr_utils.py
code served (permissive licence) · get_code("f7a82fb0484dfd06")
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 dict. 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

torch.manual_seed(0)

# Construct arguments for build_mean_dict
# The function takes structure, means, and stds
# It zips them together and builds a dictionary
# structure can contain either strings (for simple keys) or tuples (for nested keys)
# means and stds are numeric values that get formatted as strings

structure = [
    'loss',
    'accuracy',
    ('metrics', 'precision'),
    ('metrics', 'recall'),
    ('metrics', 'f1')
]

means = [0.1234, 0.8765, 0.9012, 0.8901, 0.8950]

stds = [0.0123, 0.0234, 0.0145, 0.0167, 0.0189]

ARGS = {
    'structure': structure,
    'means': means,
    'stds': stds
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
extract_mean_report Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/mean_cr_utils.py
code served (permissive licence) · get_code("d0d83d9f0237e9de")
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 numpy as np

# Create sample data for c_dicts parameter
# The function expects a list of dictionaries where values can be either:
# 1. Scalar/numeric values (handled directly)
# 2. Nested dictionaries with numeric values

c_dicts = [
    {
        "metric_a": 1.5,
        "metric_b": 2.3,
        "nested": {
            "sub_1": 0.5,
            "sub_2": 0.8
        }
    },
    {
        "metric_a": 1.7,
        "metric_b": 2.1,
        "nested": {
            "sub_1": 0.6,
            "sub_2": 0.9
        }
    },
    {
        "metric_a": 1.6,
        "metric_b": 2.2,
        "nested": {
            "sub_1": 0.55,
            "sub_2": 0.85
        }
    }
]

ARGS = {
    "c_dicts": c_dicts
}
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 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
get_datasets Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/ucr_loader.py
code served (permissive licence) · get_code("70634cdff58ce5b5")
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 dict. 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 os
import tempfile
from glob import glob

# Create a temporary directory structure to satisfy the glob pattern
tmpdir = tempfile.mkdtemp()

# Create some test files in a nested structure
# The function splits by '/')[-2] which gets the second-to-last path component
subdir1 = os.path.join(tmpdir, "category1", "subdir")
subdir2 = os.path.join(tmpdir, "category2", "subdir")
os.makedirs(subdir1, exist_ok=True)
os.makedirs(subdir2, exist_ok=True)

# Create some dummy files
open(os.path.join(subdir1, "file1.txt"), "w").close()
open(os.path.join(subdir2, "file2.txt"), "w").close()

ARGS = {
    "root": tmpdir,
    "prefix": "**/**/",
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
maybe_create_dirs Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/utils.py
code served (permissive licence) · get_code("8f07858fabba918e")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was 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 os
import tempfile

# Create a temporary directory to use as root
temp_root = tempfile.mkdtemp()

ARGS = {
    'dataset_name': 'cifar10',
    'root': temp_root,
    'dirs': ['models', 'results'],
    'exp': 'experiment_1',
    'return_paths': True,
    'verbose': 1
}
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 list, ndim 1, dtype family float — a list 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
perform_datasplit Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/utils.py
code served (permissive licence) · get_code("f17f927bb375a913")
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 numpy as np
from sklearn.model_selection import train_test_split

# Create sample data for the function
np.random.seed(0)
num_samples = 20
num_features = 4

# Create numeric data (features)
data = np.random.randn(num_samples, num_features)

# Create stratifiable labels (discrete classes)
labels = np.random.randint(0, 3, num_samples)

# Define arguments dictionary
ARGS = {
    'data': data,
    'labels': labels,
    'test_split': 0.3,
    'stratify': True,
    'return_state': False,
    'random_state': 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
pretty_print Ran
ran on a synthesized input
DominiqueMercier/PPML-TSA/src/modules/mean_cr_utils.py
code served (permissive licence) · get_code("286b204dc50eec25")
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 str. 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

# Analyze the function to understand what mean_dict should contain
# The function expects:
# - Keys that include class names (all but last 3)
# - Last 3 keys: 'accuracy', 'macro avg', 'weighted avg'
# - Each class key maps to a dict with 'precision', 'recall', 'f1-score', 'support'
# - 'accuracy' is a string
# - 'macro avg' and 'weighted avg' are dicts with those metrics

mean_dict = {
    'class_0': {
        'precision': '0.85',
        'recall': '0.90',
        'f1-score': '0.87',
        'support': '100.0'
    },
    'class_1': {
        'precision': '0.92',
        'recall': '0.88',
        'f1-score': '0.90',
        'support': '95.0'
    },
    'accuracy': '0.89',
    'macro avg': {
        'precision': '0.885',
        'recall': '0.89',
        'f1-score': '0.885',
        'support': '195.0'
    },
    'weighted avg': {
        'precision': '0.884',
        'recall': '0.89',
        'f1-score': '0.885',
        'support': '195.0'
    }
}

ARGS = {
    'mean_dict': mean_dict
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
to_array Not yet run
unverified (harvested, no recorded run)
DominiqueMercier/PPML-TSA/src/modules/ucr_loader.py
code served (permissive licence) · get_code("0c5a7b6544c1736a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

With the advent of machine learning in applications of critical infrastructure such as healthcare and energy, privacy is a growing concern in the minds of stakeholders. It is pivotal to ensure that neither the model nor the data can be used to extract sensitive information used by attackers against individuals or to harm whole societies through the exploitation of critical infrastructure. The applicability of machine learning in these domains is mostly limited due to a lack of trust regarding the transparency and the privacy constraints. Various safety-critical use cases (mostly relying on time-series data) are currently underrepresented in privacy-related considerations. By evaluating several privacy-preserving methods regarding their applicability on time-series data, we validated the inefficacy of encryption for deep learning, the strong dataset dependence of differential privacy, and the broad applicability of federated methods.

For agents

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

get_harvested_code_for_paper("2111.14838")
get_code_for_paper("2111.14838")
have("2111.14838")

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

curl https://syntology.ai/api/ran/2111.14838.json

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

[![Syntology run record](https://syntology.ai/api/ran/2111.14838.svg)](https://syntology.ai/paper/2111.14838)

Connect an agent — have() is free.