SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1602.02697 · 2016

Practical Black-Box Attacks against Machine Learning

arXiv · PDF · Open in the Atlas

Code that ran

6 of 11 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; 3 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 5 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/1602.02697.json · badge: /api/ran/1602.02697.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
probabilistic-jsmas/probabilistic-jsmas reimplementation (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
jbinas/fortified-networks community (classifier label)
classifier basis not recorded on the link
2 ran · 5 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
arg_type Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from shijiel2/cleverhans
pointer only · get_code("50b95028f0c76d76")
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.

dim2int Ran
ran on a synthesized input
jbinas/fortified-networks/fortnet_dae/resnet_tf.py
code served (permissive licence) · get_code("2250a446cd9d9750")
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 int, ndim 0, dtype family int, 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
from enum import Enum

class Dimension:
    def __init__(self, value):
        self.value = value

ARGS = {
    "dim": Dimension(5)
}
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 int, ndim 0, dtype family int, 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
norm_l0 Ran
ran, honoured the contract we drafted
probabilistic-jsmas/probabilistic-jsmas/start_stats.py
pointer only (licence: NONE) · get_code("b265d88671923e2c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, probabilistic-jsmas/probabilistic-jsmas 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.

norm_l1 Ran
ran, honoured the contract we drafted
probabilistic-jsmas/probabilistic-jsmas/start_stats.py
pointer only (licence: NONE) · get_code("3488e2e71e94a1a4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, probabilistic-jsmas/probabilistic-jsmas 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.

norm_l2 Ran
ran, honoured the contract we drafted
probabilistic-jsmas/probabilistic-jsmas/start_stats.py
pointer only (licence: NONE) · get_code("824708a335504c38")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, probabilistic-jsmas/probabilistic-jsmas 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.

weight_rec_err Ran
ran on a synthesized input
jbinas/fortified-networks/run_blackbox.py
code served (permissive licence) · get_code("3bb1443aebde61bb")
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 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

torch.manual_seed(0)

# The function weight_rec_err takes two parameters:
# - err: can be None or something that supports multiplication
# - wgt: a weight value that multiplies err

# From the function body:
# - err can be None (checked explicitly)
# - err and wgt are multiplied together, so they should support * operator
# - Both are likely scalars or tensors

err = torch.tensor(2.5)
wgt = torch.tensor(0.8)

ARGS = {
    "err": err,
    "wgt": wgt
}
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 Tensor, 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
compute_rec_err Not yet run
unverified (harvested, no recorded run)
jbinas/fortified-networks/fortnet_dae/utils_tf.py
code served (permissive licence) · get_code("9b208439f9e823d7")
How it ran

no invocation recorded.

load_mnist Not yet run
unverified (harvested, no recorded run)
jbinas/fortified-networks/datasets/utils_fashion_mnist.py
code served (permissive licence) · get_code("7802d4d8bfea87e4")
How it ran

no invocation recorded.

make_resnet_model Not yet run
unverified (harvested, no recorded run)
jbinas/fortified-networks/fortnet_dae/models_tf.py
code served (permissive licence) · get_code("8f0d60173f7c6db6")
How it ran

no invocation recorded.

model_loss Not yet run
unverified (harvested, no recorded run)
jbinas/fortified-networks/fortnet_dae/utils_tf.py
code served (permissive licence) · get_code("befd82aa2271735d")
How it ran

no invocation recorded.

preprocess_image Not yet run
unverified (harvested, no recorded run)
jbinas/fortified-networks/datasets/utils_cifar10.py
code served (permissive licence) · get_code("6d943e78fa13a160")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 later) and 9 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
arg_type 2 other papers
1 dated earlier, 1 later
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
1610.00768 (2016-10)
Technical Report on the CleverHans v2.1.0 Adversarial Examples Library
load_mnist 2 other papers
2 later
1708.07747 (2017-08)
Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
2606.04451 (2026-06)
On Out-of-sample Embedding in UMAP

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

Machine learning (ML) models, e.g., deep neural networks (DNNs), are vulnerable to adversarial examples: malicious inputs modified to yield erroneous model outputs, while appearing unmodified to human observers. Potential attacks include having malicious content like malware identified as legitimate or controlling vehicle behavior. Yet, all existing adversarial example attacks require knowledge of either the model internals or its training data. We introduce the first practical demonstration of an attacker controlling a remotely hosted DNN with no such knowledge. Indeed, the only capability of our black-box adversary is to observe labels given by the DNN to chosen inputs. Our attack strategy consists in training a local model to substitute for the target DNN, using inputs synthetically generated by an adversary and labeled by the target DNN. We use the local substitute to craft adversarial examples, and find that they are misclassified by the targeted DNN. To perform a real-world and properly-blinded evaluation, we attack a DNN hosted by MetaMind, an online deep learning API. We find that their DNN misclassifies 84.24% of the adversarial examples crafted with our substitute. We demonstrate the general applicability of our strategy to many ML techniques by conducting the same attack against models hosted by Amazon and Google, using logistic regression substitutes. They yield adversarial examples misclassified by Amazon and Google at rates of 96.19% and 88.94%. We also find that this black-box attack strategy is capable of evading defense strategies previously found to make adversarial example crafting harder.

For agents

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

get_harvested_code_for_paper("1602.02697")
get_code_for_paper("1602.02697")
have("1602.02697")

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

curl https://syntology.ai/api/ran/1602.02697.json

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

[![Syntology run record](https://syntology.ai/api/ran/1602.02697.svg)](https://syntology.ai/paper/1602.02697)

Connect an agent — have() is free.