SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.02844 · ICML · 2025

Wolfpack Adversarial Attack for Robust Multi-Agent Reinforcement Learning

Seungyul Han, Yonghyeon Jo, Sunwoo Lee, Jaebak Hwang

arXiv · PDF · Open in the Atlas

Code that ran

2 of 7 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; 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/2502.02844.json · badge: /api/ran/2502.02844.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
sunwoolee0504/wall canonical (classifier label)
classifier basis: classifier_2025
2 ran · 3 no recorded run
zzq-bot/ROMANCE community
Papers with Code archive
0 ran · 2 no recorded run
FunctionStatusWhere it lives
DMAQ_SI_Weight Ran
ran on a synthesized input
sunwoolee0504/wall/src/learners/dmaq_qatten_learner.py
code served (permissive licence) · get_code("bcce7c3814b850a4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, 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 torch as th
import numpy as np
from types import SimpleNamespace

# Create args object for __init__
args = SimpleNamespace()
args.n_agents = 3
args.n_actions = 5
args.state_shape = (12,)  # Will give state_dim = 12
args.num_kernel = 2
args.adv_hypernet_embed = 8
args.adv_hypernet_layers = 2

INIT_ARGS = {'args': args}

# Create inputs for forward pass
batch_size = 2
state_dim = int(np.prod(args.state_shape))  # 12
action_dim = args.n_agents * args.n_actions  # 3 * 5 = 15

states = th.randn(batch_size, state_dim)
actions = th.randn(batch_size, action_dim)

ARGS = {
    'states': states,
    'actions': actions
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 2, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Qatten_Weight Ran
ran on a synthesized input
sunwoolee0504/wall/src/learners/dmaq_qatten_learner.py
code served (permissive licence) · get_code("4fd93532d4a24457")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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 torch as th
import numpy as np

# Create a minimal args object with all required attributes
class Args:
    def __init__(self):
        self.n_agents = 3
        self.state_shape = (15,)  # Will give state_dim = 15
        self.unit_dim = 5  # 5 * 3 agents = 15, fits in state_dim
        self.n_actions = 4
        self.n_head = 2
        self.mixing_embed_dim = 8
        self.attend_reg_coef = 0.001
        self.hypernet_embed = 6
        self.nonlinear = True
        self.weighted_head = True
        self.mask_dead = False
        self.state_bias = True

args = Args()

INIT_ARGS = {
    'args': args
}

# Prepare inputs for forward pass
# agent_qs: Q-values for each agent, shape should be (batch, n_agents)
batch_size = 2
agent_qs = th.randn(batch_size, args.n_agents)

# states: shape (batch, state_dim) = (batch, 15)
# First unit_dim * n_agents = 5 * 3 = 15 elements are unit states
states = th.randn(batch_size, 15)

# actions: shape (batch, n_agents) - used for masking if mask_dead is True
actions = th.randint(1, args.n_actions + 1, (batch_size, args.n_agents))

ARGS = {
    'agent_qs': agent_qs,
    'states': states,
    'actions': actions
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
numpy, torch — 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
AttackMAC Not yet run
unverified (harvested, no recorded run)
zzq-bot/ROMANCE/src/controllers/attack_controller.py
code served (permissive licence) · get_code("99dcaa1bc3b17c24")
How it ran

no invocation recorded.

BasicMAC Not yet run
unverified (harvested, no recorded run)
zzq-bot/ROMANCE/src/controllers/attack_controller.py
code served (permissive licence) · get_code("081f018f212fad8a")
How it ran

no invocation recorded.

DMAQ_QattenMixer Not yet run
unverified (harvested, no recorded run)
sunwoolee0504/wall/src/learners/dmaq_qatten_learner.py
code served (permissive licence) · get_code("a924257e72cfa10c")
How it ran

no invocation recorded.

DMAQ_qattenLearner Not yet run
unverified (harvested, no recorded run)
sunwoolee0504/wall/src/learners/dmaq_qatten_learner.py
code served (permissive licence) · get_code("7d4a9f1d322d2bf6")
How it ran

no invocation recorded.

DMAQer Not yet run
unverified (harvested, no recorded run)
sunwoolee0504/wall/src/learners/dmaq_qatten_learner.py
code served (permissive licence) · get_code("f1884a61380317b4")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 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.

FunctionAlso present inEarliest-datedOthers named
DMAQ_SI_Weight
brick of this paper's assembly: helper
2 other papers
2 dated earlier
2206.11420 (2022-06)
PAC: Assisted Value Factorisation with Counterfactual Predictions in Multi-Agent Reinforcement Learning
2305.05909 (2023-05)
Robust Multi-Agent Coordination via Evolutionary Generation of Auxiliary Adversarial Attackers *
DMAQer
brick of this paper's assembly: helper
2 other papers
2 dated earlier
2206.11420 (2022-06)
PAC: Assisted Value Factorisation with Counterfactual Predictions in Multi-Agent Reinforcement Learning
2305.05909 (2023-05)
Robust Multi-Agent Coordination via Evolutionary Generation of Auxiliary Adversarial Attackers *
DMAQ_QattenMixer
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2305.05909 (2023-05)
Robust Multi-Agent Coordination via Evolutionary Generation of Auxiliary Adversarial Attackers *
Qatten_Weight
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2305.05909 (2023-05)
Robust Multi-Agent Coordination via Evolutionary Generation of Auxiliary Adversarial Attackers *

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

Traditional robust methods in multi-agent reinforcement learning (MARL) often struggle against coordinated adversarial attacks in cooperative scenarios. To address this limitation, we propose the Wolfpack Adversarial Attack framework, inspired by wolf hunting strategies, which targets an initial agent and its assisting agents to disrupt cooperation. Additionally, we introduce the Wolfpack-Adversarial Learning for MARL (WALL) framework, which trains robust MARL policies to defend against the proposed Wolfpack attack by fostering systemwide collaboration. Experimental results underscore the devastating impact of the Wolfpack attack and the significant robustness improvements achieved by WALL. Our code is available at https://github.com/sunwoolee0504/WALL.

For agents

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

get_harvested_code_for_paper("2502.02844")
get_code_for_paper("2502.02844")
have("2502.02844")

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

curl https://syntology.ai/api/ran/2502.02844.json

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

[![Syntology run record](https://syntology.ai/api/ran/2502.02844.svg)](https://syntology.ai/paper/2502.02844)

Connect an agent — have() is free.