SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.08669 · 2022

MPAF: Model Poisoning Attacks to Federated Learning based on Fake Clients

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/2203.08669.json · badge: /api/ran/2203.08669.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
yjlee22/byzantinefl community (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
FunctionStatusWhere it lives
compromised_clients Ran
ran on a synthesized input
yjlee22/byzantinefl/utils/attack.py
code served (permissive licence) · get_code("1e24d2c7e37c042a")
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 int; 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 random

class Args:
    def __init__(self, c_frac, num_clients):
        self.c_frac = c_frac
        self.num_clients = num_clients

ARGS = {
    'args': Args(c_frac=0.3, num_clients=10)
}
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 int — 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
euclid Ran
ran on a synthesized input
yjlee22/byzantinefl/utils/byzantine_fl.py
code served (permissive licence) · get_code("bd25babe9e928a3b")
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

torch.manual_seed(0)

# Create two small tensors with distinct dimensions
# v1 and v2 should be 1D vectors for matrix multiplication to work
v1 = torch.randn(4)
v2 = torch.randn(4)

ARGS = {
    "v1": v1,
    "v2": v2,
}
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
fedavg Ran
ran on a synthesized input
yjlee22/byzantinefl/src/aggregation.py
code served (permissive licence) · get_code("2ff65e8d813a99c6")
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 copy
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a list of local model weight dictionaries
# Each dictionary contains parameter tensors that would come from a federated learning client
w_locals = [
    {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
        'layer2.weight': torch.randn(3, 2),
        'layer2.bias': torch.randn(3),
    },
    {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
        'layer2.weight': torch.randn(3, 2),
        'layer2.bias': torch.randn(3),
    },
    {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
        'layer2.weight': torch.randn(3, 2),
        'layer2.bias': torch.randn(3),
    },
]

ARGS = {
    'w_locals': w_locals,
}
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
multi_vectorization Ran
ran on a synthesized input
yjlee22/byzantinefl/utils/byzantine_fl.py
code served (permissive licence) · get_code("cf3480639eff1b33")
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 2, 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 copy
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple args object with device attribute
class Args:
    def __init__(self):
        self.device = 'cpu'

args = Args()

# w_locals should be a list of dicts, where each dict maps names to tensors
# The function reshapes each tensor and concatenates them
w_locals = [
    {
        'weight': torch.randn(2, 4),
        'bias': torch.randn(3)
    },
    {
        'weight': torch.randn(5, 2),
        'bias': torch.randn(1)
    }
]

ARGS = {
    'w_locals': w_locals,
    'args': args
}
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 2, 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
noniid Ran
ran on a synthesized input
yjlee22/byzantinefl/utils/sampling.py
code served (permissive licence) · get_code("ee77c61bebb1af09")
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 numpy as np
from collections import namedtuple

# Create a minimal dataset class with labels attribute
class MinimalDataset:
    def __init__(self, num_samples, num_classes):
        # Create labels for the dataset
        self.labels = np.random.randint(0, num_classes, size=num_samples)
    
    def __len__(self):
        return len(self.labels)

# Create a minimal args class with all required attributes
class Args:
    def __init__(self):
        self.num_clients = 5
        self.num_classes = 3
        self.num_data = 100
        self.alpha = 0.5
        self.quantity_skew = True

# Set random seed for reproducibility
np.random.seed(0)

# Create dataset and args
dataset = MinimalDataset(num_samples=500, num_classes=3)
args = Args()

ARGS = {
    "dataset": dataset,
    "args": args
}
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 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
single_vectorization Ran
ran on a synthesized input
yjlee22/byzantinefl/utils/byzantine_fl.py
code served (permissive licence) · get_code("944c967ac42a6dc4")
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 (36,) 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 copy

torch.manual_seed(0)

# Create a small model state dict (w_glob parameter)
w_glob = {
    'layer1.weight': torch.randn(4, 3),
    'layer1.bias': torch.randn(4),
    'layer2.weight': torch.randn(5, 4),
}

# Create args object with device attribute
class Args:
    def __init__(self):
        self.device = 'cpu'

args = Args()

ARGS = {
    'w_glob': w_glob,
    'args': args,
}
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 (36,) — shape (36,) 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
untargeted_attack Ran
ran on a synthesized input
yjlee22/byzantinefl/utils/attack.py
code served (permissive licence) · get_code("c665d5e709dcbb19")
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 copy

torch.manual_seed(0)

# Build a small dict of tensors to represent model weights
w = {
    'layer1': torch.randn(2, 4, dtype=torch.float32),
    'layer2': torch.randn(3, 4, dtype=torch.float32),
}

# Build args object with required attributes
class Args:
    def __init__(self):
        self.device = 'cpu'
        self.mp_lambda = 0.5

args = Args()

ARGS = {
    'w': w,
    'args': args,
}
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
tabulate_events Not yet run
unverified (harvested, no recorded run)
yjlee22/byzantinefl/tb2csv.py
code served (permissive licence) · get_code("ab3c6082071f0ec1")
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

Existing model poisoning attacks to federated learning assume that an attacker has access to a large fraction of compromised genuine clients. However, such assumption is not realistic in production federated learning systems that involve millions of clients. In this work, we propose the first Model Poisoning Attack based on Fake clients called MPAF. Specifically, we assume the attacker injects fake clients to a federated learning system and sends carefully crafted fake local model updates to the cloud server during training, such that the learnt global model has low accuracy for many indiscriminate test inputs. Towards this goal, our attack drags the global model towards an attacker-chosen base model that has low accuracy. Specifically, in each round of federated learning, the fake clients craft fake local model updates that point to the base model and scale them up to amplify their impact before sending them to the cloud server. Our experiments show that MPAF can significantly decrease the test accuracy of the global model, even if classical defenses and norm clipping are adopted, highlighting the need for more advanced defenses.

For agents

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

get_harvested_code_for_paper("2203.08669")
get_code_for_paper("2203.08669")
have("2203.08669")

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

curl https://syntology.ai/api/ran/2203.08669.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.08669.svg)](https://syntology.ai/paper/2203.08669)

Connect an agent — have() is free.