SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1911.04448 · 2019

Real-Time Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

16 of 16 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 16 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/1911.04448.json · badge: /api/ran/1911.04448.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
rmst/rtrl canonical (classifier label)
classifier basis: census
8 ran · 0 no recorded run
elementai/avenue canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
yannbouteiller/gym-airsimdroneracinglab community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
FunctionStatusWhere it lives
basis_oriented_quaternion Ran
ran on a synthesized input
yannbouteiller/gym-airsimdroneracinglab/gym_game_of_drones/envs/multi_agent/gym_airsimdroneracinglab/airsimdroneracinglab_env.py
code served (permissive licence) · get_code("4f1fcee7a0959171")
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 Quaternion. 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 torch

class Quaternion:
    """Stand-in quaternion class with inverse() method and multiplication support"""
    def __init__(self, x, y, z, w):
        self.x = x
        self.y = y
        self.z = z
        self.w = w
    
    def inverse(self):
        """Return the inverse of this quaternion"""
        norm_sq = self.x**2 + self.y**2 + self.z**2 + self.w**2
        return Quaternion(
            -self.x / norm_sq,
            -self.y / norm_sq,
            -self.z / norm_sq,
            self.w / norm_sq
        )
    
    def __mul__(self, other):
        """Quaternion multiplication"""
        w1, x1, y1, z1 = self.w, self.x, self.y, self.z
        w2, x2, y2, z2 = other.w, other.x, other.y, other.z
        
        w = w1*w2 - x1*x2 - y1*y2 - z1*z2
        x = w1*x2 + x1*w2 + y1*z2 - z1*y2
        y = w1*y2 - x1*z2 + y1*w2 + z1*x2
        z = w1*z2 + x1*y2 - y1*x2 + z1*w2
        
        return Quaternion(x, y, z, w)


torch.manual_seed(0)

drone_base_orientation = Quaternion(x=0.1, y=0.2, z=0.3, w=0.9)
base_quaternion = Quaternion(x=0.05, y=0.15, z=0.25, w=0.95)

ARGS = {
    "drone_base_orientation": drone_base_orientation,
    "base_quaternion": base_quaternion
}
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 Quaternion — a Quaternion output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Quaternion 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
collate Ran
ran on a synthesized input
rmst/rtrl/rtrl/util.py
code served (permissive licence) · get_code("37983d1795ec599f")
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 3, dtype family float, shape (2, 4, 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 torch
import numpy as np
from collections.abc import Sequence, Mapping

torch.manual_seed(0)

# Test batch with torch tensors
batch_tensors = [
    torch.randn(4, 3),
    torch.randn(4, 3),
]

# Test batch with numpy arrays
batch_numpy = [
    np.random.randn(4, 3),
    np.random.randn(4, 3),
]

# Test batch with sequences (lists of tensors)
batch_sequences = [
    (torch.randn(4, 3), torch.randn(5, 2)),
    (torch.randn(4, 3), torch.randn(5, 2)),
]

# Test batch with mappings (dicts)
batch_mappings = [
    {'x': torch.randn(4, 3), 'y': torch.randn(2,)},
    {'x': torch.randn(4, 3), 'y': torch.randn(2,)},
]

# Test batch with custom objects that have __torch_tensor__
class TorchTensorWrapper:
    def __init__(self, shape):
        self._tensor = torch.randn(*shape)
    
    def __torch_tensor__(self):
        return self._tensor

batch_torch_tensor_protocol = [
    TorchTensorWrapper((4, 3)),
    TorchTensorWrapper((4, 3)),
]

# Test batch with scalars (fallback case)
batch_scalars = [1, 2, 3, 4]

ARGS = {
    'batch': batch_tensors,
    'device': None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 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
compute_assed_id Ran
ran on a synthesized input
elementai/avenue/avenue/util.py
code served (permissive licence) · get_code("9e6f271628db1153")
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.

# Setup code for compute_assed_id

ARGS = {
    'name': 'myapp',
    'system': 'linux'
}
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
compute_asset_path Ran
ran on a synthesized input
elementai/avenue/avenue/util.py
code served (permissive licence) · get_code("3898cde4a2e9842f")
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

# Set up environment variable for assets directory (optional, tests default behavior)
os.environ['AVENUE_ASSETS'] = '/tmp/test_assets'

# The function takes a single parameter: asset_id
# It's used as a string to join with directory paths
asset_id = "model_v1.onnx"

ARGS = {
    "asset_id": asset_id
}
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
concat_spaces_from_dict Ran
ran on a synthesized input
elementai/avenue/avenue/wrappers.py
code served (permissive licence) · get_code("38e629a2f8bc42de")
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 Box. 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 gym
import numpy as np

# Create gym.spaces.Box objects for the spaces dict
# The function expects spaces to have 'shape', 'low', 'high', and 'dtype' attributes
space1 = gym.spaces.Box(low=0.0, high=1.0, shape=(2, 3), dtype=np.float32)
space2 = gym.spaces.Box(low=0.0, high=1.0, shape=(2, 4), dtype=np.float32)
space3 = gym.spaces.Box(low=0.0, high=1.0, shape=(2, 5), dtype=np.float32)

spaces = {
    'obs_1': space1,
    'obs_2': space2,
    'obs_3': space3,
}

keys = ['obs_1', 'obs_2', 'obs_3']

ARGS = {
    'spaces': spaces,
    'keys': keys,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Box — a Box output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Box 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
copy_shared Ran
ran on a synthesized input
rmst/rtrl/rtrl/nn.py
code served (permissive licence) · get_code("f3ed23b35cd6dba4")
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 SimpleModel. 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 torch
import torch.nn as nn
from copy import deepcopy

torch.manual_seed(0)

# Create a simple model with some parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model_a = SimpleModel()

ARGS = {
    "model_a": model_a
}
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 SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
dqn_conv Ran
ran on a synthesized input
rmst/rtrl/rtrl/nn.py
code served (permissive licence) · get_code("c64381a3a6c81fd5")
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 torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "n": 3
}
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 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 1 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
float64_to_float32 Ran
ran on a synthesized input
rmst/rtrl/rtrl/wrappers.py
code served (permissive licence) · get_code("d4a1d5bcb45f2a7b")
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 numpy as np

# Test case 1: float64 array that should be converted to float32
x_float64 = np.array([1.5, 2.7, 3.9], dtype=np.float64)

# Test case 2: float32 array that should remain unchanged
x_float32 = np.array([1.5, 2.7, 3.9], dtype=np.float32)

# We'll use the float64 case as the main argument since it exercises the conversion
ARGS = {
    "x": x_float64
}
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
get_wrapper_by_class Ran
ran on a synthesized input
rmst/rtrl/rtrl/wrappers.py
code served (permissive licence) · get_code("49fd27673ae96565")
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 CustomWrapper. 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 gym
import torch

torch.manual_seed(0)

class DummyEnv(gym.Env):
    def __init__(self):
        self.observation_space = gym.spaces.Box(0, 1, shape=(4,))
        self.action_space = gym.spaces.Discrete(2)
    
    def reset(self):
        return self.observation_space.sample()
    
    def step(self, action):
        return self.observation_space.sample(), 0, False, {}
    
    def render(self):
        pass

class CustomWrapper(gym.Wrapper):
    def __init__(self, env):
        super().__init__(env)

class AnotherWrapper(gym.Wrapper):
    def __init__(self, env):
        super().__init__(env)

# Create a nested wrapper structure
base_env = DummyEnv()
wrapped_env = CustomWrapper(base_env)
double_wrapped_env = AnotherWrapper(wrapped_env)

# Test case 1: env is already an instance of the target class
env_test_1 = wrapped_env
cls_test_1 = CustomWrapper

ARGS = {
    "env": env_test_1,
    "cls": cls_test_1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CustomWrapper — a CustomWrapper output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CustomWrapper 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
min_max_norm Ran
ran on a synthesized input
elementai/avenue/avenue/util.py
code served (permissive licence) · get_code("bbaf3f539ccf9404")
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
import numpy as np

torch.manual_seed(0)

# Construct arguments for min_max_norm function
# The function performs min-max normalization: ((x - min_value) / (max_value - min_value) - 0.5) * 2
# It takes x (tensor/array), min_value (scalar), and max_value (scalar)

x = torch.randn(2, 4)  # Small tensor: batch size 2, dimension 4
min_value = 0.0
max_value = 1.0

ARGS = {
    "x": x,
    "min_value": min_value,
    "max_value": max_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
no_grad Ran
ran on a synthesized input
rmst/rtrl/rtrl/nn.py
code served (permissive licence) · get_code("064f68f60a5e4279")
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 SimpleModel. 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 torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
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 SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
partition Ran
ran on a synthesized input
rmst/rtrl/rtrl/util.py
code served (permissive licence) · get_code("c0504dae0e2af9cc")
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, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
from typing import Mapping

torch.manual_seed(0)

# The function partition() takes a single parameter 'x'
# It handles torch.Tensor and Mapping types
# For Mapping case: it creates a dict of partitioned values,
# then constructs tuples of new dicts with individual elements

# Test case 1: A simple Mapping (dict) with tensor values
test_dict = {
    'a': torch.randn(2, 4),
    'b': torch.randn(2, 3),
}

ARGS = {
    'x': test_dict
}
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 tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
quaternion_to_euler Ran
ran on a synthesized input
yannbouteiller/gym-airsimdroneracinglab/gym_game_of_drones/dataset_collector_game.py
code served (permissive licence) · get_code("445cd0db640c75dc")
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 numpy as np

# The function takes a quaternion q as input
# q should be array-like with 4 elements: [x, y, z, w]
q = np.array([0.0, 0.0, 0.0, 1.0])

ARGS = {
    "q": q
}
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
quaternion_to_yaw Ran
ran on a synthesized input
yannbouteiller/gym-airsimdroneracinglab/gym_game_of_drones/envs/multi_agent/gym_airsimdroneracinglab/rewardfunction.py
code served (permissive licence) · get_code("b4dafd10c6a29ce4")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

class Quaternion:
    def __init__(self, w, x, y, z):
        self.w_val = w
        self.x_val = x
        self.y_val = y
        self.z_val = z

q = Quaternion(
    w=1.0,
    x=0.0,
    y=0.5,
    z=0.3
)

ARGS = {
    "q": q
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
quaternions_to_gate_yaw_cos Ran
ran on a synthesized input
yannbouteiller/gym-airsimdroneracinglab/gym_game_of_drones/envs/multi_agent/gym_airsimdroneracinglab/rewardfunction.py
code served (permissive licence) · get_code("fa1168e9585a6ae1")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

class Quaternion:
    def __init__(self, w, x, y, z):
        self.w_val = w
        self.x_val = x
        self.y_val = y
        self.z_val = z
    
    def inverse(self):
        # For a unit quaternion, inverse is the conjugate
        norm_sq = self.w_val**2 + self.x_val**2 + self.y_val**2 + self.z_val**2
        return Quaternion(
            self.w_val / norm_sq,
            -self.x_val / norm_sq,
            -self.y_val / norm_sq,
            -self.z_val / norm_sq
        )
    
    def __mul__(self, other):
        # Quaternion multiplication
        w = self.w_val * other.w_val - self.x_val * other.x_val - self.y_val * other.y_val - self.z_val * other.z_val
        x = self.w_val * other.x_val + self.x_val * other.w_val + self.y_val * other.z_val - self.z_val * other.y_val
        y = self.w_val * other.y_val - self.x_val * other.z_val + self.y_val * other.w_val + self.z_val * other.x_val
        z = self.w_val * other.z_val + self.x_val * other.y_val - self.y_val * other.x_val + self.z_val * other.w_val
        return Quaternion(w, x, y, z)

# Create two quaternion objects
q1 = Quaternion(w=0.9238795, x=0.0, y=0.3826834, z=0.0)
q2 = Quaternion(w=0.7071068, x=0.0, y=0.7071068, z=0.0)

ARGS = {
    'q1': q1,
    'q2': q2
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
shallow_copy Ran
ran on a synthesized input
rmst/rtrl/rtrl/util.py
code served (permissive licence) · get_code("592c3fc322538b71")
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 SimpleObject. 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 torch
import numpy as np

torch.manual_seed(0)

class SimpleObject:
    """A simple test object with some attributes"""
    def __init__(self):
        self.value = 42
        self.name = "test"
        self.data = [1, 2, 3]
        self.nested = {"key": "value"}

# Create an instance to shallow copy
obj = SimpleObject()

ARGS = {
    "obj": obj
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleObject — a SimpleObject output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleObject 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

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 14 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
copy_shared 1 other paper
1 later
2010.02966 (2020-10)
Reinforcement Learning with Random Delays
no_grad 1 other paper
1 later
2010.02966 (2020-10)
Reinforcement Learning with Random Delays

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

Markov Decision Processes (MDPs), the mathematical framework underlying most algorithms in Reinforcement Learning (RL), are often used in a way that wrongfully assumes that the state of an agent's environment does not change during action selection. As RL systems based on MDPs begin to find application in real-world safety critical situations, this mismatch between the assumptions underlying classical MDPs and the reality of real-time computation may lead to undesirable outcomes. In this paper, we introduce a new framework, in which states and actions evolve simultaneously and show how it is related to the classical MDP formulation. We analyze existing algorithms under the new real-time formulation and show why they are suboptimal when used in real-time. We then use those insights to create a new algorithm Real-Time Actor-Critic (RTAC) that outperforms the existing state-of-the-art continuous control algorithm Soft Actor-Critic both in real-time and non-real-time settings. Code and videos can be found at https://github.com/rmst/rtrl.

For agents

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

get_harvested_code_for_paper("1911.04448")
get_code_for_paper("1911.04448")
have("1911.04448")

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

curl https://syntology.ai/api/ran/1911.04448.json

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

[![Syntology run record](https://syntology.ai/api/ran/1911.04448.svg)](https://syntology.ai/paper/1911.04448)

Connect an agent — have() is free.