SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1912.01588 · 2019

Leveraging Procedural Generation to Benchmark Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

12 of 15 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, 1 violated it and 4 have no contract verdict; 5 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/1912.01588.json · badge: /api/ran/1912.01588.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
openai/train-procgen canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
rgilman33/carlita community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
ana-tudor/classyconditioning community (classifier label)
classifier basis not recorded on the link
3 ran · 2 no recorded run
joonleesky/train-procgen-pytorch reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
constfn Ran
ran; our contract draft was wrong, not the code
ana-tudor/classyconditioning/fruitbot_ppo/ppo_agent.py
pointer only (licence: MIT) · get_code("8758308dc6c0727f")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, ana-tudor/classyconditioning, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

drop_connect Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rgilman33/carlita/efficientnet_utils.py
pointer only (licence: MIT) · get_code("d3319e3d34ca90ca")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, rgilman33/carlita, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

ema Ran
ran on a synthesized input
openai/train-procgen/train_procgen/graph_util.py
code served (permissive licence) · get_code("a362c435a091c0f4")
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 ndarray, ndim 1, dtype family float, shape (5,) 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 numpy as np

# Create sample input data for the EMA function
# The function expects a 1D or multi-dimensional array of numerical values
data_in = np.array([1.0, 2.0, 3.0, 4.0, 5.0], dtype=np.float32)

# Define the ARGS dictionary with all parameters
ARGS = {
    "data_in": data_in,
    "smoothing": 0.3
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (5,) — shape (5,) 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
get_hidden Ran
ran on a synthesized input
rgilman33/carlita/train_utils.py
pointer only (licence: MIT) · get_code("3a87e9cb29c5eff4")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, rgilman33/carlita, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

read_csv Ran
ran on a synthesized input
openai/train-procgen/train_procgen/graph_util.py
code served (permissive licence) · get_code("b5e9129d63af4909")
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 ndarray, ndim 1, dtype family float, shape (3,) 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 csv
import tempfile
import os
import numpy as np

# Create a temporary CSV file for testing
temp_dir = tempfile.mkdtemp()
csv_filename = os.path.join(temp_dir, "test_data.csv")

# Write a sample CSV file with a header row and data rows
with open(csv_filename, 'w', newline='') as f:
    writer = csv.writer(f)
    writer.writerow(['Name', 'Score', 'Value'])
    writer.writerow(['Alice', '95.5', '10.2'])
    writer.writerow(['Bob', '87.3', '20.5'])
    writer.writerow(['Charlie', '92.1', '15.8'])

ARGS = {
    'filename': csv_filename,
    'key_name': 'score'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (3,) 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
round_filters Ran
ran; our contract draft was wrong, not the code
rgilman33/carlita/efficientnet_utils.py
pointer only (licence: MIT) · get_code("f15a49337e69e937")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, rgilman33/carlita, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

round_repeats Ran
ran; our contract draft was wrong, not the code
rgilman33/carlita/efficientnet_utils.py
pointer only (licence: MIT) · get_code("dbc0ca08d119a5a0")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, rgilman33/carlita, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

run Ran
ran; our contract draft was wrong, not the code
rgilman33/carlita/procgen/build.py
pointer only (licence: MIT) · get_code("1067e655bdab66ee")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, rgilman33/carlita, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

safemean Ran
ran, honoured the contract we drafted
ana-tudor/classyconditioning/fruitbot_ppo/ppo_agent.py
pointer only (licence: MIT) · get_code("a7b323241cd70612")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, ana-tudor/classyconditioning, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

sf01 Ran
ran, violated the contract we drafted
ana-tudor/classyconditioning/fruitbot_ppo/epopt_runner.py
pointer only (licence: MIT) · get_code("baf2a28d376d7803")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, ana-tudor/classyconditioning, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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.

switch_to_outer_plot Ran
ran on a synthesized input
openai/train-procgen/train_procgen/graph_util.py
code served (permissive licence) · get_code("a080ad6d9c26a9e9")
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 Axes. 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 matplotlib.pyplot as plt

fig = plt.figure()

ARGS = {
    "fig": fig
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Axes — a Axes output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Axes 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
update_mean_var_count_from_moments Ran
ran, honoured the contract we drafted
joonleesky/train-procgen-pytorch/common/env/procgen_wrappers.py
pointer only (licence: NONE) · get_code("3ebb420a5d10cc64")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, joonleesky/train-procgen-pytorch 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.

build_reg_impala_cnn Not yet run
unverified (harvested, no recorded run)
ana-tudor/classyconditioning/fruitbot_ppo/reg_impala_cnn.py
pointer only (licence: MIT) · get_code("5d6006f62f7660ef")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, ana-tudor/classyconditioning, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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. No invocation recorded.

build_reg_impala_cnn_verbose Not yet run
unverified (harvested, no recorded run)
ana-tudor/classyconditioning/fruitbot_ppo/reg_impala_cnn.py
pointer only (licence: MIT) · get_code("f9d33adb8b6ca4d2")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, ana-tudor/classyconditioning, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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. No invocation recorded.

get_var Not yet run
unverified (harvested, no recorded run)
rgilman33/carlita/procgen-build/procgen_build/build_package.py
pointer only (licence: MIT); the stored copy get_code serves, from JacobPfau/procgenAISC, is pointer only (licence: MIT) · get_code("fa2053a8568916ec")
How it ran

The body of this sample is withheld: MIT, as GitHub detects it on this paper's repository, rgilman33/carlita, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with 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. No invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 11 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 2 the same month, 11 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
round_filters 31 other papers
3 dated earlier, 1 the same month, 27 later · 3 only as an assembly brick
1808.08127 (2018-08)
Recalibrating Fully Convolutional Networks with Spatial and Channel 'Squeeze & Excitation' Blocks
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
1911.09070 (2019-11)
EfficientDet: Scalable and Efficient Object Detection
and 28 more not named here
round_repeats 30 other papers
3 dated earlier, 1 the same month, 26 later · 3 only as an assembly brick
1808.08127 (2018-08)
Recalibrating Fully Convolutional Networks with Spatial and Channel 'Squeeze & Excitation' Blocks
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
1911.09070 (2019-11)
EfficientDet: Scalable and Efficient Object Detection
and 27 more not named here
drop_connect 23 other papers
1 dated earlier, 22 later · 2 only as an assembly brick
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
2010.00475 (2020-10)
FSD50K: An Open Dataset of Human-Labeled Sound Events
2107.01105 (2021-07)
Memory Efficient Meta-Learning with Large Images
and 20 more not named here
constfn 8 other papers
7 dated earlier, 1 later
1707.06347 (2017-07)
Proximal Policy Optimization Algorithms
1806.07857 (2018-06)
RUDDER: Return Decomposition for Delayed Rewards
1812.02341 (2018-12)
Quantifying Generalization in Reinforcement Learning
and 5 more not named here
sf01 8 other papers
7 dated earlier, 1 later
1707.06347 (2017-07)
Proximal Policy Optimization Algorithms
1806.07857 (2018-06)
RUDDER: Return Decomposition for Delayed Rewards
1812.02341 (2018-12)
Quantifying Generalization in Reinforcement Learning
and 5 more not named here
update_mean_var_count_from_moments 8 other papers
1 dated earlier, 7 later · 1 only as an assembly brick
1901.09184 (2019-01)
Action Robust Reinforcement Learning and Applications in Continuous Control
2006.14171 (2020-06)
A Closer Look at Invalid Action Masking in Policy Gradient Algorithms
2105.10682 (2021-05)
Feasible Actor-Critic: Constrained Reinforcement Learning for Ensuring Statewise Safety
and 5 more not named here
safemean 6 other papers
4 dated earlier, 2 later · 1 only as an assembly brick
1707.06347 (2017-07)
Proximal Policy Optimization Algorithms
1804.06459 (2018-04)
On Learning Intrinsic Rewards for Policy Gradient Methods
1905.02363 (2019-05)
Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning
and 3 more not named here
run 3 other papers
3 later
2105.14111 (2021-05)
Goal Misgeneralization in Deep Reinforcement Learning
2411.03008 (2024-11)
Hierarchical Orchestra of Policies
2605.10546 (2026-05)
Higher Resolution, Better Generalization: Unlocking Visual Scaling in Deep Reinforcement Learning
ema 1 other paper
1 later
2009.04416 (2020-09)
Phasic Policy Gradient
get_var 1 other paper
1 later
2105.14111 (2021-05)
Goal Misgeneralization in Deep Reinforcement Learning
switch_to_outer_plot 1 other paper
1 later
2009.04416 (2020-09)
Phasic Policy Gradient

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

We introduce Procgen Benchmark, a suite of 16 procedurally generated game-like environments designed to benchmark both sample efficiency and generalization in reinforcement learning. We believe that the community will benefit from increased access to high quality training environments, and we provide detailed experimental protocols for using this benchmark. We empirically demonstrate that diverse environment distributions are essential to adequately train and evaluate RL agents, thereby motivating the extensive use of procedural content generation. We then use this benchmark to investigate the effects of scaling model size, finding that larger models significantly improve both sample efficiency and generalization.

For agents

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

get_harvested_code_for_paper("1912.01588")
get_code_for_paper("1912.01588")
have("1912.01588")

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

curl https://syntology.ai/api/ran/1912.01588.json

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

[![Syntology run record](https://syntology.ai/api/ran/1912.01588.svg)](https://syntology.ai/paper/1912.01588)

Connect an agent — have() is free.