SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.11229 · 2021

Off-Policy Correction For Multi-Agent Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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 3 have no contract verdict; 0 failures of our instrument; 0 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/2111.11229.json · badge: /api/ran/2111.11229.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
awarelab/seed_rl canonical (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
FunctionStatusWhere it lives
create_mlp Ran
ran on a synthesized input
awarelab/seed_rl/agents/sac/networks.py
code served (permissive licence) · get_code("4a1477f42fca643a")
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 Sequential. 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 tensorflow as tf

# Define the arguments for create_mlp
hidden_sizes = [8, 16, 4]
hidden_activation = 'relu'
last_layer_activation = 'sigmoid'

ARGS = {
    'hidden_sizes': hidden_sizes,
    'hidden_activation': hidden_activation,
    'last_layer_activation': last_layer_activation
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential 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
initial_frame_stacking_state Ran
ran on a synthesized input
awarelab/seed_rl/atari/networks.py
code served (permissive licence) · get_code("8de5cad377fd4152")
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 EagerTensor, ndim 2, dtype family int, shape (4, 7056) 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 tensorflow as tf

ARGS = {
    "stack_size": 2,
    "batch_size": tf.constant(4, dtype=tf.int32),
    "observation_shape": [84, 84, 1],
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family int, shape (4, 7056) — shape (4, 7056) 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
stack_frames Ran
ran on a synthesized input
awarelab/seed_rl/atari/networks.py
code served (permissive licence) · get_code("bfc70838af9ecd54")
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 tensorflow as tf
import numpy as np

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

# Define parameters for stack_frames
time_steps = 3
batch_size = 2
height = 4
width = 3
channels = 1
stack_size = 3

# Create frames tensor: <float32>[time, batch_size, height, width, channels]
# Range [0, 255]
frames = tf.constant(
    np.random.randint(0, 256, size=(time_steps, batch_size, height, width, channels)),
    dtype=tf.float32
)

# Create frame_stacking_state tensor: <int32>[batch_size, height*width]
# Bit-packed frames in range [0, 256)
obs_elements = height * width
frame_stacking_state = tf.constant(
    np.random.randint(0, 256, size=(batch_size, obs_elements)),
    dtype=tf.int32
)

# Create done tensor: <bool>[time, batch_size]
done = tf.constant(
    np.random.randint(0, 2, size=(time_steps, batch_size), dtype=bool),
    dtype=tf.bool
)

# stack_size is an integer
# Already defined above as 3

ARGS = {
    "frames": frames,
    "frame_stacking_state": frame_stacking_state,
    "done": done,
    "stack_size": stack_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 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
create_mlp 1 other paper
1 dated earlier
1910.06591 (2019-10)
SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference
initial_frame_stacking_state 1 other paper
1 dated earlier
1910.06591 (2019-10)
SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference
stack_frames 1 other paper
1 dated earlier
1910.06591 (2019-10)
SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference

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) provides a framework for problems involving multiple interacting agents. Despite apparent similarity to the single-agent case, multi-agent problems are often harder to train and analyze theoretically. In this work, we propose MA-Trace, a new on-policy actor-critic algorithm, which extends V-Trace to the MARL setting. The key advantage of our algorithm is its high scalability in a multi-worker setting. To this end, MA-Trace utilizes importance sampling as an off-policy correction method, which allows distributing the computations with no impact on the quality of training. Furthermore, our algorithm is theoretically grounded - we prove a fixed-point theorem that guarantees convergence. We evaluate the algorithm extensively on the StarCraft Multi-Agent Challenge, a standard benchmark for multi-agent algorithms. MA-Trace achieves high performance on all its tasks and exceeds state-of-the-art results on some of them.

For agents

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

get_harvested_code_for_paper("2111.11229")
get_code_for_paper("2111.11229")
have("2111.11229")

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

curl https://syntology.ai/api/ran/2111.11229.json

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

[![Syntology run record](https://syntology.ai/api/ran/2111.11229.svg)](https://syntology.ai/paper/2111.11229)

Connect an agent — have() is free.