SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.08432 · 2020

Data Poisoning Attacks Against Federated Learning Systems

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 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 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/2007.08432.json · badge: /api/ran/2007.08432.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
mattbev/6.867-final-project community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
FunctionStatusWhere it lives
generic_train Ran
ran on a synthesized input
mattbev/6.867-final-project/utils/basics.py
code served (permissive licence) · get_code("ac98353e654a66d7")
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 torch
import torch.nn as nn
import torch.optim as optim

torch.manual_seed(0)

# Define a minimal model
class SimpleModel(nn.Module):
    def __init__(self):
        super(SimpleModel, self).__init__()
        self.fc = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.fc(x)

# Define a minimal attack class
class MinimalAttack:
    def train(self, model, trainloader):
        pass
    
    def run(self, inputs, labels):
        return inputs, labels

# Create model
model = SimpleModel()

# Create a minimal dataloader (list of batches)
trainloader = [
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
    (torch.randn(2, 4), torch.randint(0, 3, (2,)))
]

# Create optimizer
optimizer = optim.SGD(model.parameters(), lr=0.01)

# Create loss criterion
criterion = nn.CrossEntropyLoss()

# Create attack
attack = MinimalAttack()

ARGS = {
    'model': model,
    'num_epochs': 2,
    'trainloader': trainloader,
    'optimizer': optimizer,
    'criterion': criterion,
    'attack': attack,
    'device': 'cpu',
    'verbose': False
}
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 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
merge Ran
ran on a synthesized input
mattbev/6.867-final-project/utils/basics.py
code served (permissive licence) · get_code("6eba28e32c29abf0")
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 3, dtype family float, shape (8, 3, 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

# Construct valid arguments for merge function
# The function expects:
# - images: numpy array with shape (batch, height, width, channels)
# - size: tuple of (rows, cols) for grid layout

# Create a small batch of images with shape (batch, height, width, channels)
# Using distinct sizes: batch=2, height=4, width=3, channels=3
images = np.random.rand(2, 4, 3, 3).astype(np.float32)

# size parameter: (2, 1) means 2 rows and 1 column in the merged grid
# This will arrange 2 images in a 2x1 grid
size = (2, 1)

ARGS = {
    "images": images,
    "size": size
}
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 3, dtype family float, shape (8, 3, 3) — shape (8, 3, 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
imsave Not yet run
unverified (harvested, no recorded run)
mattbev/6.867-final-project/utils/basics.py
code served (permissive licence) · get_code("ff180d9ec05d76f9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 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.

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

Federated learning (FL) is an emerging paradigm for distributed training of large-scale deep neural networks in which participants' data remains on their own devices with only model updates being shared with a central server. However, the distributed nature of FL gives rise to new threats caused by potentially malicious participants. In this paper, we study targeted data poisoning attacks against FL systems in which a malicious subset of the participants aim to poison the global model by sending model updates derived from mislabeled data. We first demonstrate that such data poisoning attacks can cause substantial drops in classification accuracy and recall, even with a small percentage of malicious participants. We additionally show that the attacks can be targeted, i.e., they have a large negative impact only on classes that are under attack. We also study attack longevity in early/late round training, the impact of malicious participant availability, and the relationships between the two. Finally, we propose a defense strategy that can help identify malicious participants in FL to circumvent poisoning attacks, and demonstrate its effectiveness.

For agents

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

get_harvested_code_for_paper("2007.08432")
get_code_for_paper("2007.08432")
have("2007.08432")

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

curl https://syntology.ai/api/ran/2007.08432.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.08432.svg)](https://syntology.ai/paper/2007.08432)

Connect an agent — have() is free.