SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.09092 · 2022

Indiscriminate Data Poisoning Attacks on Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

5 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; 3 honoured the contract we drafted, 0 violated it and 2 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/2204.09092.json · badge: /api/ran/2204.09092.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
watml/tgda-attack canonical (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
FunctionStatusWhere it lives
autograd Ran
ran, honoured the contract we drafted
watml/tgda-attack/loss_function.py
pointer only (licence: MIT); the stored copy get_code serves, from RufengXiao/AdaRHD, is pointer only (licence: NONE) · get_code("59a762d8a2476e97")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from RufengXiao/AdaRHD, and GitHub detected no licence in that repository 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.

concat Ran
ran on a synthesized input
watml/tgda-attack/utils.py
code served (permissive licence) · get_code("cc489bfd1fb0f0d4")
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. 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 two small lists of tensors
tensors_one = [
    torch.randn(2, 4),
    torch.randn(2, 3),
]

tensors_two = [
    torch.randn(2, 5),
    torch.randn(2, 2),
]

ARGS = {
    "tensors_one": tensors_one,
    "tensors_two": tensors_two,
}
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 list — a list 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
dot Ran
ran, honoured the contract we drafted
watml/tgda-attack/utils.py
code served (permissive licence) · get_code("33077b2243c0f0a0")
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 1, dtype family float, shape (1,) 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 takes two parameters: tensors_one and tensors_two
# It iterates through them with zip and performs element-wise multiplication and sum
# Both should be iterables (lists) of tensors with compatible shapes

tensors_one = [
    torch.randn(2, 4),
    torch.randn(2, 4),
    torch.randn(3, 5)
]

tensors_two = [
    torch.randn(2, 4),
    torch.randn(2, 4),
    torch.randn(3, 5)
]

ARGS = {
    "tensors_one": tensors_one,
    "tensors_two": tensors_two
}
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 1, dtype family float, shape (1,) — shape (1,) 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
leader_loss Ran
ran on a synthesized input
watml/tgda-attack/loss_function.py
code served (permissive licence) · get_code("7a6c24c8f0b4e54c")
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 torch.nn as nn

torch.manual_seed(0)

# Define a simple follower model that takes real_data and outputs logits
class SimpleFollower(nn.Module):
    def __init__(self, input_dim=4, num_classes=3):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.linear(x)

# Define a simple leader model (not used in the function but required as parameter)
class SimpleLeader(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create instances
leader = SimpleLeader()
follower = SimpleFollower(input_dim=4, num_classes=3)

# Create real_data: batch_size=2, feature_dim=4
real_data = torch.randn(2, 4)

# Create real_y: batch_size=2 with class labels (0, 1, or 2 for 3 classes)
real_y = torch.tensor([0, 2])

ARGS = {
    "leader": leader,
    "follower": follower,
    "real_data": real_data,
    "real_y": real_y
}
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 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
norm Ran
ran, honoured the contract we drafted
watml/tgda-attack/utils.py
code served (permissive licence) · get_code("02aa687cd06876bd")
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
import torch

torch.manual_seed(0)

# The function takes a single parameter 'tensors' which should be an iterable
# of tensors. The function iterates through them, squares each element,
# sums them, and computes the square root of the total sum.

# Create a list of small tensors with distinct shapes
tensor1 = torch.randn(2, 4)
tensor2 = torch.randn(3, 5)
tensor3 = torch.randn(2, 3)

tensors = [tensor1, tensor2, tensor3]

ARGS = {
    "tensors": tensors
}
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 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
follower_loss Not yet run
unverified (harvested, no recorded run)
watml/tgda-attack/loss_function.py
code served (permissive licence) · get_code("efd73a00d518db9e")
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, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 3 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
autograd 2 other papers
2 later
2402.03883 (2024-02)
A Framework for Bilevel Optimization on Riemannian Manifolds
2504.06042 (2025-04)
An Adaptive Algorithm for Bilevel Optimization on Riemannian Manifolds
dot 2 other papers
2 later
2402.03883 (2024-02)
A Framework for Bilevel Optimization on Riemannian Manifolds
2504.06042 (2025-04)
An Adaptive Algorithm for Bilevel Optimization on Riemannian Manifolds
norm 1 other paper
1 later
2402.03883 (2024-02)
A Framework for Bilevel Optimization on Riemannian Manifolds

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 poisoning attacks, in which a malicious adversary aims to influence a model by injecting "poisoned" data into the training process, have attracted significant recent attention. In this work, we take a closer look at existing poisoning attacks and connect them with old and new algorithms for solving sequential Stackelberg games. By choosing an appropriate loss function for the attacker and optimizing with algorithms that exploit second-order information, we design poisoning attacks that are effective on neural networks. We present efficient implementations that exploit modern auto-differentiation packages and allow simultaneous and coordinated generation of tens of thousands of poisoned points, in contrast to existing methods that generate poisoned points one by one. We further perform extensive experiments that empirically explore the effect of data poisoning attacks on deep neural networks.

For agents

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

get_harvested_code_for_paper("2204.09092")
get_code_for_paper("2204.09092")
have("2204.09092")

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

curl https://syntology.ai/api/ran/2204.09092.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.09092.svg)](https://syntology.ai/paper/2204.09092)

Connect an agent — have() is free.