SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1903.04527 · 2019

Multi-Agent Deep Reinforcement Learning for Large-scale Traffic Signal Control

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 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 5 have no contract verdict; 2 failures of our instrument; 2 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/1903.04527.json · badge: /api/ran/1903.04527.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
cts198859/deeprl_signal_control canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
FunctionStatusWhere it lives
check_dir Ran
ran on a synthesized input
cts198859/deeprl_signal_control/utils.py
code served (permissive licence) · get_code("a774e08f7631887e")
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 bool, 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 os
import tempfile

# Create a temporary directory that exists
temp_dir = tempfile.mkdtemp()

# Create ARGS dict with the cur_dir parameter
ARGS = {
    "cur_dir": temp_dir
}
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 bool, 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
create_multiplexer Ran
ran on a synthesized input
cts198859/deeprl_signal_control/extract_tensorboard.py
code served (permissive licence) · get_code("bd2260a92d6fcfef")
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 EventMultiplexer. 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 tempfile
import os
from tensorboard.backend.event_processing import plugin_event_multiplexer as event_multiplexer
from tensorboard.compat.proto import summary_pb2
from tensorboard.compat.proto import event_pb2
import time

# Create a temporary directory to serve as run_path
temp_dir = tempfile.mkdtemp()

# Create a minimal event file in the temporary directory
event_file_path = os.path.join(temp_dir, "events.out.tfevents.0.hostname")

# Write a minimal event file with a scalar summary
writer = event_pb2.Event()
writer.wall_time = time.time()
writer.file_version = "brain.Event:2"

with open(event_file_path, 'wb') as f:
    f.write(writer.SerializeToString())
    
    # Write a scalar event
    scalar_event = event_pb2.Event()
    scalar_event.wall_time = time.time()
    scalar_event.step = 1
    
    summary = summary_pb2.Summary()
    value = summary.value.add()
    value.tag = "loss"
    value.simple_value = 0.5
    scalar_event.summary.CopyFrom(summary)
    
    f.write(scalar_event.SerializeToString())

run_name = "test_run"
run_path = temp_dir

ARGS = {
    "run_path": run_path,
    "run_name": run_name,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorboard — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EventMultiplexer — a EventMultiplexer output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a EventMultiplexer 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
extract_scalar Ran
ran on a synthesized input
cts198859/deeprl_signal_control/extract_tensorboard.py
code served (permissive licence) · get_code("2a1c3fc637c91a31")
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 DataFrame. 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 pandas as pd
import tensorflow as tf
from unittest.mock import Mock

# Create a mock event class
class MockEvent:
    def __init__(self, wall_time, step, value):
        self.wall_time = wall_time
        self.step = step
        # Create a tensor_proto with the value
        self.tensor_proto = tf.make_tensor_proto([value], dtype=tf.float32)

# Create a mock multiplexer
class MockMultiplexer:
    def __init__(self):
        self.events_data = {}
    
    def Tensors(self, run_name, tag):
        key = (run_name, tag)
        if key not in self.events_data:
            self.events_data[key] = []
        return self.events_data[key]

# Set up multiplexer with some test data
multiplexer = MockMultiplexer()
run_name = "train"
tag = "loss"

# Add some mock events to the multiplexer
events = [
    MockEvent(wall_time=1.0, step=0, value=0.5),
    MockEvent(wall_time=2.0, step=1, value=0.3),
    MockEvent(wall_time=3.0, step=2, value=0.1),
]
multiplexer.events_data[(run_name, tag)] = events

ARGS = {
    'multiplexer': multiplexer,
    'run_name': run_name,
    'tag': tag,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
find_file Ran
ran; our contract draft was wrong, not the code
cts198859/deeprl_signal_control/utils.py
code served (permissive licence) · get_code("b2fb59622dad3ce4")
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 str. 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 tempfile
import os

# Create a temporary directory with a test .ini file
temp_dir = tempfile.mkdtemp()
test_ini_file = os.path.join(temp_dir, 'config.ini')
with open(test_ini_file, 'w') as f:
    f.write('[test]\nkey=value\n')

ARGS = {
    'cur_dir': temp_dir,
    'suffix': '.ini'
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
get_event_log Ran
ran on a synthesized input
cts198859/deeprl_signal_control/extract_tensorboard.py
code served (permissive licence) · get_code("eeb0e1e913b454fe")
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 str. 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 os
import tempfile
import shutil

# Create a temporary directory with a mock TensorBoard events file
temp_dir = tempfile.mkdtemp()
events_file = os.path.join(temp_dir, 'events.out.tfevents.1234567890')
with open(events_file, 'w') as f:
    f.write('mock event data')

ARGS = {
    'log_dir': temp_dir + os.sep
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
init_dir Ran
ran on a synthesized input
cts198859/deeprl_signal_control/utils.py
code served (permissive licence) · get_code("636e940d9b81af2a")
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 os
import tempfile

# Create a temporary base directory for testing
base_dir = tempfile.mkdtemp()

ARGS = {
    'base_dir': base_dir,
    'pathes': ['log', 'data', 'model']
}
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 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
ortho_init Ran
ran; our contract draft was wrong, not the code
cts198859/deeprl_signal_control/agents/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from cts198859/deeprl_network, is pointer only (licence: NONE) · get_code("6583a667fee3e221")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from cts198859/deeprl_network, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

conv Not yet run
unverified (harvested, no recorded run)
cts198859/deeprl_signal_control/agents/utils.py
code served (permissive licence) · get_code("3d1a6d68723115b4")
How it ran

no invocation recorded.

norm_init Not yet run
unverified (harvested, no recorded run)
cts198859/deeprl_signal_control/agents/utils.py
code served (permissive licence) · get_code("5af19b29d5176250")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 5 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.

FunctionAlso present inEarliest-datedOthers named
ortho_init 2 other papers
2 later
2004.01339 (2020-04)
Multi-agent Reinforcement Learning for Networked System Control
2010.07777 (2020-10)
A game-theoretic analysis of networked system control for common-pool resource management using multi-agent reinforcement learning
check_dir 1 other paper
1 later
2010.07777 (2020-10)
A game-theoretic analysis of networked system control for common-pool resource management using multi-agent reinforcement learning
conv 1 other paper
1 later
2010.07777 (2020-10)
A game-theoretic analysis of networked system control for common-pool resource management using multi-agent reinforcement learning
find_file 1 other paper
1 later
2010.07777 (2020-10)
A game-theoretic analysis of networked system control for common-pool resource management using multi-agent reinforcement learning
norm_init 1 other paper
1 later
2010.07777 (2020-10)
A game-theoretic analysis of networked system control for common-pool resource management using multi-agent reinforcement learning

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

Reinforcement learning (RL) is a promising data-driven approach for adaptive traffic signal control (ATSC) in complex urban traffic networks, and deep neural networks further enhance its learning power. However, centralized RL is infeasible for large-scale ATSC due to the extremely high dimension of the joint action space. Multi-agent RL (MARL) overcomes the scalability issue by distributing the global control to each local RL agent, but it introduces new challenges: now the environment becomes partially observable from the viewpoint of each local agent due to limited communication among agents. Most existing studies in MARL focus on designing efficient communication and coordination among traditional Q-learning agents. This paper presents, for the first time, a fully scalable and decentralized MARL algorithm for the state-of-the-art deep RL agent: advantage actor critic (A2C), within the context of ATSC. In particular, two methods are proposed to stabilize the learning procedure, by improving the observability and reducing the learning difficulty of each local agent. The proposed multi-agent A2C is compared against independent A2C and independent Q-learning algorithms, in both a large synthetic traffic grid and a large real-world traffic network of Monaco city, under simulated peak-hour traffic dynamics. Results demonstrate its optimality, robustness, and sample efficiency over other state-of-the-art decentralized MARL algorithms.

For agents

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

get_harvested_code_for_paper("1903.04527")
get_code_for_paper("1903.04527")
have("1903.04527")

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

curl https://syntology.ai/api/ran/1903.04527.json

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

[![Syntology run record](https://syntology.ai/api/ran/1903.04527.svg)](https://syntology.ai/paper/1903.04527)

Connect an agent — have() is free.