SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.05331 · 2022

Effects of Spectral Normalization in Multi-agent Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

4 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 0 no recorded run), measured 2026-11-21; 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/2212.05331.json · badge: /api/ran/2212.05331.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
kinalmehta/epymarl_spectral canonical (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
FunctionStatusWhere it lives
gumbel_softmax_sample Ran
ran; our contract draft was wrong, not the code
kinalmehta/epymarl_spectral/src_spectral/controllers/maddpg_controller.py
code served (permissive licence) · get_code("355d7b51c7ffd042")
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 2, dtype family float, shape (2, 4) 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 as th
from torch.autograd import Variable
import torch.nn.functional as F

th.manual_seed(0)

# Create a small logits tensor (batch_size=2, num_classes=4)
logits = th.randn(2, 4, requires_grad=True)

# Set temperature
temperature = 0.5

# Construct ARGS dict
ARGS = {
    "logits": logits,
    "temperature": temperature,
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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
onehot_from_logits Ran
ran; our contract draft was wrong, not the code
kinalmehta/epymarl_spectral/src_spectral/controllers/maddpg_controller.py
code served (permissive licence) · get_code("046b92243aae4f76")
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 2, dtype family float, shape (2, 4) 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 expects:
# - logits: a tensor that supports .max(-1, keepdim=True) and comparison operations
# - eps: a float (default 0.0, not used in the function body)

# Create a small batch of logits
# Shape: (batch_size=2, num_actions=4)
logits = torch.randn(2, 4)

ARGS = {
    "logits": logits,
    "eps": 0.0
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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
sample_gumbel Ran
ran; our contract draft was wrong, not the code
kinalmehta/epymarl_spectral/src_spectral/controllers/maddpg_controller.py
code served (permissive licence) · get_code("3695d66ab681cacf")
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 2, dtype family float, shape (2, 4) 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 as th
from torch.autograd import Variable

th.manual_seed(0)

# Define arguments for sample_gumbel function
shape = (2, 4)  # Small tensor shape: batch size 2, dimension 4
eps = 1e-20  # Default epsilon value
tens_type = th.FloatTensor  # Tensor type

ARGS = {
    "shape": shape,
    "eps": eps,
    "tens_type": tens_type
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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
work Ran
ran, honoured the contract we drafted
kinalmehta/epymarl_spectral/src_spectral/search.py
code served (permissive licence) · get_code("930812c3b06fc858")
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 int, ndim 0, dtype family int, 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 subprocess

def work(cmd):
    cmd = cmd.split(' ')
    return subprocess.call(cmd, shell=False)

# The function takes a cmd string, splits it by spaces, and calls subprocess.call
# We need to provide a valid command that can be executed
# Using 'echo' with a simple message as a safe, universal command

ARGS = {
    'cmd': 'echo hello'
}
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 int, ndim 0, dtype family int, 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

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, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 4 later) and 0 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
onehot_from_logits 6 other papers
3 dated earlier, 3 later
2006.07869 (2020-06)
Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Tasks
2006.09447 (2020-06)
Agent Modelling under Partial Observability for Deep Reinforcement Learning
2210.06835 (2022-10)
Multi-agent Dynamic Algorithm Configuration
and 3 more not named here
gumbel_softmax_sample 5 other papers
2 dated earlier, 3 later
2006.07869 (2020-06)
Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Tasks
2210.06835 (2022-10)
Multi-agent Dynamic Algorithm Configuration
2410.03997 (2024-10)
YOLO-MARL: You Only LLM Once for Multi-Agent Reinforcement Learning
and 2 more not named here
sample_gumbel 5 other papers
2 dated earlier, 3 later
2006.07869 (2020-06)
Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Tasks
2210.06835 (2022-10)
Multi-agent Dynamic Algorithm Configuration
2410.03997 (2024-10)
YOLO-MARL: You Only LLM Once for Multi-Agent Reinforcement Learning
and 2 more not named here
work 5 other papers
2 dated earlier, 3 later
1910.07483 (2019-10)
MAVEN: Multi-Agent Variational Exploration
2006.07869 (2020-06)
Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Tasks
2410.03997 (2024-10)
YOLO-MARL: You Only LLM Once for Multi-Agent Reinforcement Learning
and 2 more not named here

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

A reliable critic is central to on-policy actor-critic learning. But it becomes challenging to learn a reliable critic in a multi-agent sparse reward scenario due to two factors: 1) The joint action space grows exponentially with the number of agents 2) This, combined with the reward sparseness and environment noise, leads to large sample requirements for accurate learning. We show that regularising the critic with spectral normalization (SN) enables it to learn more robustly, even in multi-agent on-policy sparse reward scenarios. Our experiments show that the regularised critic is quickly able to learn from the sparse rewarding experience in the complex SMAC and RWARE domains. These findings highlight the importance of regularisation in the critic for stable learning.

For agents

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

get_harvested_code_for_paper("2212.05331")
get_code_for_paper("2212.05331")
have("2212.05331")

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

curl https://syntology.ai/api/ran/2212.05331.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.05331.svg)](https://syntology.ai/paper/2212.05331)

Connect an agent — have() is free.