SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2209.10485 · NeurIPS · 2022

Towards a Standardised Performance Evaluation Protocol for Cooperative MARL

Arnu Pretorius, Rihab Gorsane, Omayma Mahjoub, Ruan De Kock, Roland Dubb, Siddarth Singh, A D P G V D N C O M A Iq L Q P L E X M A P P O Q T R A N

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 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 0 have no contract verdict; 3 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/2209.10485.json · badge: /api/ran/2209.10485.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
instadeepai/marl-eval community
Papers with Code archive
3 ran · 1 no recorded run
FunctionStatusWhere it lives
check_comma_in_algo_names Ran
ran; our contract draft was wrong, not the code
instadeepai/marl-eval/marl_eval/utils/data_processing_utils.py
code served (permissive licence) · get_code("ef74477c720f3bec")
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 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
import numpy as np

# Setup for check_comma_in_algo_names function
# This function checks if algorithm names contain commas

# Test case 1: Mix of valid and invalid algorithm names
algos = [
    "algorithm_1",
    "algorithm,2",
    "algo_three",
    "algo,four,invalid",
    "valid_algo_5"
]

ARGS = {
    "algos": algos,
    "get_valid": False
}
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 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: none by the harness; any seed is in the setup block
data_process_pipeline Ran
ran; our contract draft was wrong, not the code
instadeepai/marl-eval/marl_eval/utils/data_processing_utils.py
code served (permissive licence) · get_code("eda68a9d12eb7e6c")
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) 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
from typing import Any, Dict, List, Tuple, Union
import numpy as np

# Construct minimal valid arguments for data_process_pipeline

# Create a small raw_data dictionary structure matching what the function expects
raw_data = {
    "environment1": {
        "task1": {
            "algorithm1": {
                "run1": {
                    "step0": {
                        "metric1": [0.5, 0.6, 0.7],
                        "metric2": [0.1, 0.2, 0.3]
                    },
                    "step1": {
                        "metric1": [0.55, 0.65, 0.75],
                        "metric2": [0.15, 0.25, 0.35]
                    }
                },
                "run2": {
                    "step0": {
                        "metric1": [0.52, 0.62, 0.72],
                        "metric2": [0.12, 0.22, 0.32]
                    },
                    "step1": {
                        "metric1": [0.57, 0.67, 0.77],
                        "metric2": [0.17, 0.27, 0.37]
                    }
                }
            },
            "algorithm2": {
                "run1": {
                    "step0": {
                        "metric1": [0.4, 0.5, 0.6],
                        "metric2": [0.2, 0.3, 0.4]
                    },
                    "step1": {
                        "metric1": [0.45, 0.55, 0.65],
                        "metric2": [0.25, 0.35, 0.45]
                    }
                }
            }
        }
    }
}

# List of metrics to normalize
metrics_to_normalize = ["metric1", "metric2"]

ARGS = {
    "raw_data": raw_data,
    "metrics_to_normalize": metrics_to_normalize
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
lower_case_dictionary_keys Ran
ran; our contract draft was wrong, not the code
instadeepai/marl-eval/marl_eval/utils/data_processing_utils.py
code served (permissive licence) · get_code("7d32b5690ab52952")
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
from typing import Any, Dict

# Create a nested dictionary with mixed case keys and various value types
nested_dict = {
    "OuterKey1": {
        "InnerKeyA": "value_a",
        "InnerKeyB": 42,
        "InnerKeyC": 3.14
    },
    "OuterKey2": {
        "InnerKeyX": [1, 2, 3],
        "InnerKeyY": {"DeepKey": "deep_value"}
    },
    "OuterKey3": {
        "MixedCase": True,
        "ALLCAPS": None
    }
}

ARGS = {
    "dictionary": nested_dict
}
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
lower_case_inputs Not yet run
unverified (harvested, no recorded run)
instadeepai/marl-eval/marl_eval/utils/data_processing_utils.py
code served (permissive licence) · get_code("5e501c3503d155e9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 4 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

Multi-agent reinforcement learning (MARL) has emerged as a useful approach to solving decentralised decision-making problems at scale. Research in the field has been growing steadily with many breakthrough algorithms proposed in recent years. In this work, we take a closer look at this rapid development with a focus on evaluation methodologies employed across a large body of research in cooperative MARL. By conducting a detailed meta-analysis of prior work, spanning 75 papers accepted for publication from 2016 to 2022, we bring to light worrying trends that put into question the true rate of progress. We further consider these trends in a wider context and take inspiration from single-agent RL literature on similar issues with recommendations that remain applicable to MARL. Combining these recommendations, with novel insights from our analysis, we propose a standardised performance evaluation protocol for cooperative MARL. We argue that such a standard protocol, if widely adopted, would greatly improve the validity and credibility of future research, make replication and reproducibility easier, as well as improve the ability of the field to accurately gauge the rate of progress over time by being able to make sound comparisons across different works. Finally, we release our meta-analysis data publicly on our project website for future research on evaluation: https://sites.google.com/view/marl-standard-protocol

For agents

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

get_harvested_code_for_paper("2209.10485")
get_code_for_paper("2209.10485")
have("2209.10485")

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

curl https://syntology.ai/api/ran/2209.10485.json

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

[![Syntology run record](https://syntology.ai/api/ran/2209.10485.svg)](https://syntology.ai/paper/2209.10485)

Connect an agent — have() is free.