SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1910.03552 · 2019

TorchBeast: A PyTorch Platform for Distributed RL

arXiv · PDF · Open in the Atlas

Code that ran

6 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 3 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/1910.03552.json · badge: /api/ran/1910.03552.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
heiner/scalable_agent canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
copy not recorded community
not in the paper's declared repositories
3 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_baseline_loss Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from aypan17/reward-misspecification
pointer only · get_code("e9123707cba84b2c")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

compute_baseline_loss Ran
ran, honoured the contract we drafted
heiner/scalable_agent/experiment.py
code served (permissive licence) · get_code("777b5a676e76d593")
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 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 tensorflow as tf
import numpy as np

# Create a small tensor for advantages
# The function computes 0.5 * sum(advantages^2), so we just need a tensor
advantages = tf.constant([1.0, 2.0, 3.0, 4.0], dtype=tf.float32)

ARGS = {
    "advantages": advantages
}
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 EagerTensor, 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
compute_entropy_loss Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from aypan17/reward-misspecification
pointer only · get_code("a27a9dc97eefda7f")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

compute_entropy_loss Ran
ran, honoured the contract we drafted
heiner/scalable_agent/experiment.py
code served (permissive licence) · get_code("50c20c55f1b07c8e")
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 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 tensorflow as tf
import numpy as np

# Create a small batch of logits
# Shape: (batch_size=2, num_classes=4)
logits = tf.constant([
    [1.0, 2.0, 0.5, -1.0],
    [0.5, -0.5, 1.5, 2.0]
], dtype=tf.float32)

ARGS = {
    "logits": logits
}
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 EagerTensor, 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
compute_policy_gradient_loss Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from aypan17/reward-misspecification
pointer only · get_code("d6d7e0906f1c57b2")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

log_probs_from_logits_and_actions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
heiner/scalable_agent/vtrace.py
code served (permissive licence) · get_code("a75de4dc4c61d7dc")
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 EagerTensor, ndim 2, dtype family float, shape (3, 2) 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
import numpy as np

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

# Define dimensions
T = 3  # time dimension
B = 2  # batch size
NUM_ACTIONS = 4  # number of actions

# Create policy_logits: shape [T, B, NUM_ACTIONS]
policy_logits = tf.random.normal(shape=[T, B, NUM_ACTIONS], dtype=tf.float32)

# Create actions: shape [T, B] with valid action indices (0 to NUM_ACTIONS-1)
actions = tf.constant(
    np.random.randint(0, NUM_ACTIONS, size=[T, B]),
    dtype=tf.int32
)

ARGS = {
    "policy_logits": policy_logits,
    "actions": actions,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, 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 float, shape (3, 2) — shape (3, 2) 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
compute_human_normalized_score Not yet run
unverified (harvested, no recorded run)
heiner/scalable_agent/dmlab30.py
code served (permissive licence) · get_code("be48a80c545e0430")
How it ran

no invocation recorded.

from_importance_weights Not yet run
unverified (harvested, no recorded run)
heiner/scalable_agent/vtrace.py
code served (permissive licence) · get_code("e11ee068514828e6")
How it ran

no invocation recorded.

from_logits Not yet run
unverified (harvested, no recorded run)
heiner/scalable_agent/vtrace.py
code served (permissive licence) · get_code("1175c77e601077b8")
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, 8 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 0 the same month, 3 later) and 1 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
compute_baseline_loss 6 other papers
2 dated earlier, 4 later
1708.04782 (2017-08)
StarCraft II: A New Challenge for Reinforcement Learning
1809.04474 (2018-09)
Multi-task Deep Reinforcement Learning with PopArt
2006.13760 (2020-06)
The NetHack Learning Environment
and 3 more not named here
compute_entropy_loss 5 other papers
1 dated earlier, 4 later
1809.04474 (2018-09)
Multi-task Deep Reinforcement Learning with PopArt
2006.13760 (2020-06)
The NetHack Learning Environment
2201.03544 (2022-01)
The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
and 2 more not named here
compute_policy_gradient_loss 2 other papers
2 later
2201.03544 (2022-01)
The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
2210.00066 (2022-10)
Improving Policy Learning via Language Dynamics Distillation
compute_baseline_loss 1 other paper
1 dated earlier
1903.01292 (2019-03)
The StreetLearn Environment and Dataset
compute_entropy_loss 1 other paper
1 dated earlier
1903.01292 (2019-03)
The StreetLearn Environment and Dataset
from_importance_weights 1 other paper
1 dated earlier
1802.01561 (2018-02)
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
from_logits 1 other paper
1 dated earlier
1802.01561 (2018-02)
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
log_probs_from_logits_and_actions 1 other paper
1 dated earlier
1802.01561 (2018-02)
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures

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

TorchBeast is a platform for reinforcement learning (RL) research in PyTorch. It implements a version of the popular IMPALA algorithm for fast, asynchronous, parallel training of RL agents. Additionally, TorchBeast has simplicity as an explicit design goal: We provide both a pure-Python implementation ("MonoBeast") as well as a multi-machine high-performance version ("PolyBeast"). In the latter, parts of the implementation are written in C++, but all parts pertaining to machine learning are kept in simple Python using PyTorch, with the environments provided using the OpenAI Gym interface. This enables researchers to conduct scalable RL research using TorchBeast without any programming knowledge beyond Python and PyTorch. In this paper, we describe the TorchBeast design principles and implementation and demonstrate that it performs on-par with IMPALA on Atari. TorchBeast is released as an open-source package under the Apache 2.0 license and is available at \url{https://github.com/facebookresearch/torchbeast}.

For agents

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

get_harvested_code_for_paper("1910.03552")
get_code_for_paper("1910.03552")
have("1910.03552")

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

curl https://syntology.ai/api/ran/1910.03552.json

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

[![Syntology run record](https://syntology.ai/api/ran/1910.03552.svg)](https://syntology.ai/paper/1910.03552)

Connect an agent — have() is free.