SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.04779 · 2020

Conservative Q-Learning for Offline Reinforcement Learning

U Berkeley, Sergey Levine, Aviral Kumar, George Tucker, Aurick Zhou

arXiv · PDF · Open in the Atlas

Code that ran

31 of 34 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (5 of them constructed an object rather than computing a result; 5 honoured the contract we drafted, 3 violated it and 8 have no contract verdict; 15 failures of our instrument; 3 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-21; 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/2006.04779.json · badge: /api/ran/2006.04779.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
young-geng/cql community (classifier label)
classifier basis not recorded on the link
12 ran · 0 no recorded run
ikostrikov/cql-results community (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
young-geng/jaxcql community (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
BY571/CQL community
Papers with Code archive
3 ran · 0 no recorded run
mehdimashayekhi/Some-RL-Implementation community
Papers with Code archive
2 ran · 0 no recorded run
runamu/compositional-conservatism community
Papers with Code archive
0 ran · 2 no recorded run
corl-team/CORL community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
flatten Ran
ran, honoured the contract we drafted
young-geng/cql/viskit/core.py
code served (permissive licence) · get_code("20922c79a0929bbf")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T22:50:02Z). 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
import numpy as np

torch.manual_seed(0)

# The function flatten takes a list of lists and flattens it into a single list
# It iterates through sublists and extracts items from each

# Create a simple nested list structure
nested_list = [
    [1, 2, 3],
    [4, 5],
    [6, 7, 8, 9]
]

ARGS = {
    "l": nested_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list 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
Actor Ran
ran on a synthesized input · constructed an object; no result was computed
BY571/CQL/CQL-SAC/agent.py
pointer only (licence: NONE) · get_code("34ba8aa41bb7fa58")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, BY571/CQL 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.

CQLAgent Ran
ran on a synthesized input · constructed an object; no result was computed
mehdimashayekhi/Some-RL-Implementation/CQL/agent.py
pointer only (licence: NONE) · get_code("3c29dab689543a89")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mehdimashayekhi/Some-RL-Implementation 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.

CQLSAC Ran
ran on a synthesized input · constructed an object; no result was computed
BY571/CQL/CQL-SAC/agent.py
pointer only (licence: NONE) · get_code("ea10c0c769c4b484")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, BY571/CQL 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.

Critic Ran
ran on a synthesized input · constructed an object; no result was computed
BY571/CQL/CQL-SAC/agent.py
pointer only (licence: NONE) · get_code("00012db4f3753643")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, BY571/CQL 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.

DDQN Ran
ran on a synthesized input · constructed an object; no result was computed
mehdimashayekhi/Some-RL-Implementation/CQL/agent.py
pointer only (licence: NONE) · get_code("82827c5a9a2325d9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mehdimashayekhi/Some-RL-Implementation 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.

batch_to_torch Ran
ran; our contract draft was wrong, not the code
young-geng/cql/SimpleSAC/replay_buffer.py
pointer only (licence: MIT); the stored copy get_code serves, from yassineCh/PGS, is pointer only (licence: NONE) · get_code("32231080b4b40efc")
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 yassineCh/PGS, 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.

clip_magnitude Ran
ran, honoured the contract we drafted
ikostrikov/cql-results/d4rl/rlkit/pythonplusplus.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aviralkumar2907/CQL, is pointer only (licence: NONE) · get_code("93d28ca67303e04b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aviralkumar2907/CQL, 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.

create_exp_name Ran
ran; our contract draft was wrong, not the code
young-geng/cql/viskit/logging.py
code served (permissive licence) · get_code("3743c5d43e639ab9")
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 datetime
import dateutil.tz

ARGS = {
    'exp_prefix': 'my_experiment',
    'exp_id': 42,
    'seed': 123,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
dateutil — 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
create_stats_ordered_dict Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ikostrikov/cql-results/d4rl/rlkit/core/eval_util.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aviralkumar2907/CQL, is pointer only (licence: NONE) · get_code("fadf5f762ff04cd6")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aviralkumar2907/CQL, 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.

dict_to_safe_json Ran
ran; our contract draft was wrong, not the code
young-geng/cql/viskit/logging.py
code served (permissive licence) · get_code("41279dfcda8b39e4")
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 torch
import numpy as np

torch.manual_seed(0)

# Test case 1: Dictionary with safe JSON values
test_dict_safe = {
    "name": "test",
    "count": 42,
    "value": 3.14,
    "flag": True,
    "empty": None,
    "list": [1, 2, 3],
    "nested_list": [[1, 2], [3, 4]],
}

# Test case 2: Dictionary with unsafe values that need conversion
tensor = torch.randn(2, 4)
numpy_array = np.array([1, 2, 3])
custom_obj = object()

test_dict_unsafe = {
    "tensor": tensor,
    "numpy_array": numpy_array,
    "custom_object": custom_obj,
    "safe_value": 123,
}

# Test case 3: Dictionary with nested dictionaries
test_dict_nested = {
    "level1": {
        "level2": {
            "value": 99,
            "tensor": torch.randn(2, 4),
        },
        "safe": "string",
    },
    "top_level": 42,
}

# Test case 4: Complex mixed dictionary
test_dict_complex = {
    "string": "hello",
    "integer": 100,
    "float": 2.718,
    "boolean": False,
    "none": None,
    "list_mixed": [1, "two", 3.0, True, None],
    "nested_dict": {
        "inner_tensor": torch.randn(3, 5),
        "inner_safe": [1, 2, 3],
    },
    "numpy": np.array([4, 5, 6]),
}

ARGS = {
    "d": test_dict_complex,
}
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 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
extend_and_repeat Ran
ran; our contract draft was wrong, not the code
young-geng/cql/SimpleSAC/model.py
pointer only (licence: MIT); the stored copy get_code serves, from zexusun/oilca-neurips23, is pointer only (licence: NONE) · get_code("1e5f5f635bdf773a")
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 zexusun/oilca-neurips23, 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.

extend_and_repeat Ran
ran; our contract draft was wrong, not the code
young-geng/jaxcql/JaxCQL/jax_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from MaxSobolMark/OOO, is pointer only (licence: NONE) · get_code("d2f199c948f2a9ed")
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 MaxSobolMark/OOO, 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.

flatten Ran
ran on a synthesized input
young-geng/jaxcql/viskit/frontend.py
code served (permissive licence) · get_code("2be75b7ac1ba23d4")
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 list, ndim 1, dtype family int; 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
import numpy as np

torch.manual_seed(0)

# The function flatten takes xs and flattens one level of nesting
# It expects xs to be an iterable of iterables
# Each inner iterable y will be iterated over to extract x

# Create a simple nested list structure
xs = [
    [1, 2, 3],
    [4, 5],
    [6, 7, 8, 9]
]

ARGS = {
    "xs": xs
}
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 list, ndim 1, dtype family int — a list 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
get_average_returns Ran
ran, honoured the contract we drafted
ikostrikov/cql-results/d4rl/rlkit/core/eval_util.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aviralkumar2907/CQL, is pointer only (licence: NONE) · get_code("a5d10a0b704c8db7")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aviralkumar2907/CQL, 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.

get_flat_params Ran
ran, honoured the contract we drafted
ikostrikov/cql-results/d4rl/rlkit/core/batch_rl_algorithm.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aviralkumar2907/CQL, is pointer only (licence: NONE) · get_code("9b9a51e3e3be7a6e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aviralkumar2907/CQL, 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.

identity Ran
ran, honoured the contract we drafted
ikostrikov/cql-results/d4rl/rlkit/pythonplusplus.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from dholzmueller/universal_double_descent, is pointer only (licence: NONE) · get_code("9910e2fc297f8665")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from dholzmueller/universal_double_descent, 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.

index_batch Ran
ran; our contract draft was wrong, not the code
young-geng/cql/SimpleSAC/replay_buffer.py
pointer only (licence: MIT); the stored copy get_code serves, from yassineCh/PGS, is pointer only (licence: NONE) · get_code("c8b2fc80a3fc7424")
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 yassineCh/PGS, 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.

is_numeric Ran
ran, violated the contract we drafted
ikostrikov/cql-results/d4rl/rlkit/pythonplusplus.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aviralkumar2907/CQL, is pointer only (licence: NONE) · get_code("4138766e854db009")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aviralkumar2907/CQL, 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.

load_progress Ran
ran; our contract draft was wrong, not the code
young-geng/cql/viskit/core.py
code served (permissive licence) · get_code("965e75a1245dd833")
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 csv
import tempfile
import os

# Create a temporary CSV file with some test data
temp_dir = tempfile.mkdtemp()
csv_path = os.path.join(temp_dir, "progress.csv")

with open(csv_path, 'w', newline='') as f:
    writer = csv.writer(f)
    writer.writerow(['epoch', 'loss', 'accuracy'])
    writer.writerow(['1', '0.5', '0.8'])
    writer.writerow(['2', '0.3', '0.9'])
    writer.writerow(['3', '0.2', '0.95'])

ARGS = {
    'progress_csv_path': csv_path
}
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 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
mse_loss Ran
ran, violated the contract we drafted
young-geng/jaxcql/JaxCQL/jax_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from MaxSobolMark/OOO, is pointer only (licence: NONE) · get_code("cc35af8b18b7dc58")
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 MaxSobolMark/OOO, 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.

multiple_action_q_function Ran
ran; our contract draft was wrong, not the code
young-geng/cql/SimpleSAC/model.py
pointer only (licence: MIT); the stored copy get_code serves, from yassineCh/PGS, is pointer only (licence: NONE) · get_code("d06ff25d52ad20f5")
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 yassineCh/PGS, 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.

multiple_action_q_function Ran
ran; our contract draft was wrong, not the code
young-geng/jaxcql/JaxCQL/model.py
code served (permissive licence) · get_code("18c566cc009ddb07")
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 function. 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)

def extend_and_repeat(tensor, axis, repeats):
    """Helper function that mimics extend_and_repeat behavior"""
    shape = list(tensor.shape)
    shape.insert(axis, repeats)
    return tensor.unsqueeze(axis).expand(shape)

class SimpleQNetwork(nn.Module):
    """Minimal Q-network that takes observations and actions"""
    def __init__(self, obs_dim=4, action_dim=3):
        super().__init__()
        self.linear = nn.Linear(obs_dim + action_dim, 1)
    
    def forward(self, observations, actions, **kwargs):
        """Compute Q-values from observations and actions"""
        combined = torch.cat([observations, actions], dim=-1)
        q_values = self.linear(combined).squeeze(-1)
        return q_values

# Create decorator function
def multiple_action_q_function(forward):
    def wrapped(self, observations, actions, **kwargs):
        multiple_actions = False
        batch_size = observations.shape[0]
        if actions.ndim == 3 and observations.ndim == 2:
            multiple_actions = True
            observations = extend_and_repeat(observations, 1, actions.shape[1]).reshape(-1, observations.shape[-1])
            actions = actions.reshape(-1, actions.shape[-1])
        q_values = forward(self, observations, actions, **kwargs)
        if multiple_actions:
            q_values = q_values.reshape(batch_size, -1)
        return q_values
    return wrapped

# Create a decorated Q-network
class DecoratedQNetwork(SimpleQNetwork):
    @multiple_action_q_function
    def forward(self, observations, actions, **kwargs):
        combined = torch.cat([observations, actions], dim=-1)
        q_values = self.linear(combined).squeeze(-1)
        return q_values

# Build test arguments
model = DecoratedQNetwork(obs_dim=4, action_dim=3)
observations = torch.randn(2, 4)
actions = torch.randn(2, 5, 3)

ARGS = {
    "forward": model.forward
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
parition_batch_train_test Ran
ran; our contract draft was wrong, not the code
young-geng/jaxcql/JaxCQL/replay_buffer.py
pointer only (licence: MIT); the stored copy get_code serves, from yassineCh/PGS, is pointer only (licence: NONE) · get_code("3f55f8769dd45c20")
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 yassineCh/PGS, 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.

prefix_metrics Ran
ran on a synthesized input
young-geng/cql/SimpleSAC/utils.py
code served (permissive licence) · get_code("d97e89c7f1953b3a")
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 torch

torch.manual_seed(0)

# Simple metrics dict with string keys and numeric values
metrics = {
    'loss': 0.5,
    'accuracy': 0.95,
    'f1_score': 0.88
}

# Prefix string to prepend to each metric key
prefix = 'train'

ARGS = {
    'metrics': metrics,
    'prefix': prefix
}
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 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
safe_json Ran
ran, violated the contract we drafted
young-geng/cql/viskit/logging.py
code served (permissive licence) · get_code("07afff96928bd142")
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 torch
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - It takes one parameter: `data`
# - It recursively checks if data is JSON-safe (None, bool, int, float, tuple, list, or dict with string keys)
# - It returns a boolean

# Test cases for a comprehensive argument:
# We'll create a complex nested structure that is JSON-safe

ARGS = {
    "data": {
        "string_key": "value",
        "number": 42,
        "float_num": 3.14,
        "bool_val": True,
        "none_val": None,
        "list": [1, 2, 3, "four"],
        "tuple": (5, 6, 7),
        "nested_dict": {
            "inner_string": "text",
            "inner_list": [True, False, None],
        },
        "complex_list": [
            {"a": 1, "b": 2},
            [1, 2, [3, 4]],
            "string",
            42,
        ],
    }
}
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 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
simple_separated_format Ran
ran; our contract draft was wrong, not the code
ikostrikov/cql-results/d4rl/rlkit/core/tabulate.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aviralkumar2907/CQL, is pointer only (licence: NONE) · get_code("5cb0c55f58279f59")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aviralkumar2907/CQL, 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.

sliding_mean Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
young-geng/cql/viskit/frontend.py
code served (permissive licence) · get_code("40f81b1d5f51ea76")
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 ndarray, ndim 1, dtype family float, shape (8,) 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

# Construct a small data array for testing
data_array = np.array([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0])

# Define the arguments dictionary
ARGS = {
    "data_array": data_array,
    "window": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 (8,) — shape (8,) 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
unique Ran
ran; our contract draft was wrong, not the code
young-geng/cql/viskit/core.py
code served (permissive licence) · get_code("6c7a4a9349a395e9")
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 list, ndim 1, dtype family int; 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
import numpy as np

torch.manual_seed(0)

# The function `unique(l)` takes a list and returns unique elements
# It converts the list to a set and back to a list
# So we need to pass a list with some duplicate elements

ARGS = {
    "l": [1, 2, 3, 2, 1, 4, 3, 5]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list 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
update_target_network Ran
ran; our contract draft was wrong, not the code
young-geng/jaxcql/JaxCQL/model.py
code served (permissive licence) · get_code("f23c0c9faf00d830")
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 jax
import jax.numpy as jnp

# Create nested parameter structures using JAX arrays
# These will simulate main_params and target_params in a neural network context

# Create a simple nested structure of parameters (like what a neural network might have)
main_params = {
    'layer1': {
        'w': jnp.array([[1.0, 2.0, 3.0, 4.0],
                        [5.0, 6.0, 7.0, 8.0],
                        [9.0, 10.0, 11.0, 12.0]]),
        'b': jnp.array([0.1, 0.2, 0.3, 0.4])
    },
    'layer2': {
        'w': jnp.array([[1.5, 2.5],
                        [3.5, 4.5],
                        [5.5, 6.5],
                        [7.5, 8.5]]),
        'b': jnp.array([0.5, 0.6])
    }
}

target_params = {
    'layer1': {
        'w': jnp.array([[0.5, 1.0, 1.5, 2.0],
                        [2.5, 3.0, 3.5, 4.0],
                        [4.5, 5.0, 5.5, 6.0]]),
        'b': jnp.array([0.01, 0.02, 0.03, 0.04])
    },
    'layer2': {
        'w': jnp.array([[0.75, 1.25],
                        [1.75, 2.25],
                        [2.75, 3.25],
                        [3.75, 4.25]]),
        'b': jnp.array([0.25, 0.35])
    }
}

tau = 0.01

ARGS = {
    'main_params': main_params,
    'target_params': target_params,
    'tau': tau
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — 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
value_and_multi_grad Ran
ran on a synthesized input
young-geng/jaxcql/JaxCQL/jax_utils.py
code served (permissive licence) · get_code("6f4355a6f1314166")
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 function. 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 jax
import jax.numpy as jnp

# Define a simple function that returns multiple outputs
def multi_output_fun(x, y):
    """A function that returns multiple outputs"""
    output1 = x * 2 + y
    output2 = x * 3 - y
    output3 = x + y * 4
    return jnp.stack([output1, output2, output3])

# Alternative version with auxiliary output
def multi_output_fun_with_aux(x, y):
    """A function that returns multiple outputs plus auxiliary info"""
    output1 = x * 2 + y
    output2 = x * 3 - y
    output3 = x + y * 4
    outputs = jnp.stack([output1, output2, output3])
    aux = {"metadata": "auxiliary_data"}
    return outputs, aux

# Create test input arrays
x_test = jnp.array([1.0, 2.0, 3.0, 4.0])
y_test = jnp.array([0.5, 1.5, 2.5, 3.5])

# Build ARGS for the basic case (no auxiliary output)
ARGS = {
    "fun": multi_output_fun,
    "n_outputs": 3,
    "argnums": 0,
    "has_aux": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
CQLPolicy Not yet run
unverified (harvested, no recorded run)
runamu/compositional-conservatism/offlinerlkit/policy/model_free/cql.py
code served (permissive licence) · get_code("badd95b06c97467f")
How it ran

no invocation recorded.

ContinuousCQL Not yet run
unverified (harvested, no recorded run)
corl-team/CORL/algorithms/offline/cql.py
code served (permissive licence) · get_code("68d38c5b86b74f05")
How it ran

no invocation recorded.

SACPolicy Not yet run
unverified (harvested, no recorded run)
runamu/compositional-conservatism/offlinerlkit/policy/model_free/cql.py
code served (permissive licence) · get_code("3487f84da3586b8f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 34 samples lifted from this paper's linked repositories, 34 carry a function identity and 0 do not. Of the 33 distinct functions with an identity, 25 are also present in other papers whose code we hold (9 with at least one paper dated earlier, 7 the same month, 25 later) and 8 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
identity 69 other papers
14 dated earlier, 3 the same month, 52 later · 5 only as an assembly brick
1706.07296 (2017-06)
A Minimal Developmental Model Can Increase Evolvability in Soft Robots
1802.09419 (2018-02)
Stochastic Hyperparameter Optimization through Hypernetworks
1804.06055 (2018-04)
Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation
and 66 more not named here
simple_separated_format 27 other papers
5 dated earlier, 2 the same month, 20 later
1903.08254 (2019-03)
Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables
1910.11670 (2019-10)
Contextual Imagined Goals for Self-Supervised Robotic Learning
1910.12827 (2019-10)
Entity Abstraction in Visual Model-Based Reinforcement Learning
and 24 more not named here
get_average_returns 24 other papers
3 dated earlier, 1 the same month, 20 later
1903.08254 (2019-03)
Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables
1910.11670 (2019-10)
Contextual Imagined Goals for Self-Supervised Robotic Learning
1912.11032 (2019-12)
Towards Practical Multi-Object Manipulation using Relational Reinforcement Learning
and 21 more not named here
flatten 21 other papers
6 dated earlier, 1 the same month, 14 later
1609.08675 (2016-09)
YouTube-8M: A Large-Scale Video Classification Benchmark
1802.04821 (2018-02)
Evolved Policy Gradients
1811.05014 (2018-11)
NeXtVLAD: An Efficient Neural Network to Aggregate Frame-level Features for Large-scale Video Classification
and 18 more not named here
create_stats_ordered_dict 18 other papers
1 dated earlier, 1 the same month, 16 later
1910.11670 (2019-10)
Contextual Imagined Goals for Self-Supervised Robotic Learning
2006.13916 (2020-06)
Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifiers
2010.08252 (2020-10)
Hyperparameter Auto-Tuning in Self-Supervised Robotic Learning
and 15 more not named here
clip_magnitude 15 other papers
4 dated earlier, 1 the same month, 10 later
1910.11670 (2019-10)
Contextual Imagined Goals for Self-Supervised Robotic Learning
1910.12827 (2019-10)
Entity Abstraction in Visual Model-Based Reinforcement Learning
1911.08453 (2019-11)
Planning with Goal-Conditioned Policies
and 12 more not named here
is_numeric 14 other papers
4 dated earlier, 1 the same month, 9 later
1910.11670 (2019-10)
Contextual Imagined Goals for Self-Supervised Robotic Learning
1910.12827 (2019-10)
Entity Abstraction in Visual Model-Based Reinforcement Learning
1911.08453 (2019-11)
Planning with Goal-Conditioned Policies
and 11 more not named here
dict_to_safe_json 6 other papers
6 later
2007.09070 (2020-07)
Hybrid Discriminative-Generative Training via Contrastive Learning
2210.07484 (2022-10)
Mutual Information Regularized Offline Reinforcement Learning
2405.17098 (2024-05)
Q-value Regularized Transformer for Offline Reinforcement Learning
and 3 more not named here
safe_json 5 other papers
5 later
2210.07484 (2022-10)
Mutual Information Regularized Offline Reinforcement Learning
2405.17098 (2024-05)
Q-value Regularized Transformer for Offline Reinforcement Learning
2405.19690 (2024-05)
Diffusion Policies Creating a Trust Region for Offline Reinforcement Learning
and 2 more not named here
unique 5 other papers
1 dated earlier, 4 later
1802.04821 (2018-02)
Evolved Policy Gradients
2007.09070 (2020-07)
Hybrid Discriminative-Generative Training via Contrastive Learning
2210.06518 (2022-10)
Semi-Supervised Offline Reinforcement Learning with Action-Free Trajectories
and 2 more not named here
extend_and_repeat 4 other papers
4 later
2210.07484 (2022-10)
Mutual Information Regularized Offline Reinforcement Learning
2210.13431 (2022-10)
Instruction-Following Agents with Multimodal Transformer
2309.10790 (2023-09)
Guide Your Agent with Adaptive Multimodal Rewards
and 1 more not named here
load_progress 4 other papers
4 later
2007.09070 (2020-07)
Hybrid Discriminative-Generative Training via Contrastive Learning
2210.06518 (2022-10)
Semi-Supervised Offline Reinforcement Learning with Action-Free Trajectories
2301.12842 (2023-01)
Direct Preference-based Policy Optimization without Reward Modeling
and 1 more not named here
mse_loss 4 other papers
4 later
2210.07484 (2022-10)
Mutual Information Regularized Offline Reinforcement Learning
2210.13431 (2022-10)
Instruction-Following Agents with Multimodal Transformer
2309.10790 (2023-09)
Guide Your Agent with Adaptive Multimodal Rewards
and 1 more not named here
multiple_action_q_function 4 other papers
4 later
2301.12842 (2023-01)
Direct Preference-based Policy Optimization without Reward Modeling
2303.00957 (2023-03)
Preference Transformer: Modeling Human Preferences using Transformers for RL
2310.17966 (2023-10)
Train Once, Get a Family: State-Adaptive Balances for Offline-to-Online Reinforcement Learning
and 1 more not named here
extend_and_repeat 3 other papers
3 later
2207.10050 (2022-07)
Discriminator-Weighted Offline Imitation Learning from Suboptimal Demonstrations
2210.06518 (2022-10)
Semi-Supervised Offline Reinforcement Learning with Action-Free Trajectories
2405.05349 (2024-05)
Offline Model-Based Optimization via Policy-Guided Gradient Search
get_flat_params 3 other papers
3 later
2311.05067 (2023-11)
Accelerating Exploration with Unlabeled Prior Data
2605.14779 (2026-05)
PENG'S Q(λ) FOR CONSERVATIVE VALUE ESTIMATION IN OFFLINE REINFORCEMENT LEARNING
2608.02332 (2026-08)
Diffusion Policy with Behavioral Advantage Correction for Offline Reinforcement Learning
update_target_network 3 other papers
3 later
2310.08558 (2023-10)
Offline Retraining for Online RL: Decoupled Policy Learning to Mitigate Exploration Bias
2310.17966 (2023-10)
Train Once, Get a Family: State-Adaptive Balances for Offline-to-Online Reinforcement Learning
2410.18076 (2024-10)
Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration
batch_to_torch 2 other papers
2 later
2210.06518 (2022-10)
Semi-Supervised Offline Reinforcement Learning with Action-Free Trajectories
2405.05349 (2024-05)
Offline Model-Based Optimization via Policy-Guided Gradient Search
index_batch 2 other papers
2 later
2210.07484 (2022-10)
Mutual Information Regularized Offline Reinforcement Learning
2405.05349 (2024-05)
Offline Model-Based Optimization via Policy-Guided Gradient Search
parition_batch_train_test 2 other papers
2 later
2210.07484 (2022-10)
Mutual Information Regularized Offline Reinforcement Learning
2405.05349 (2024-05)
Offline Model-Based Optimization via Policy-Guided Gradient Search

Shown: the 20 functions present in the most other papers. 5 more shared functions are not listed here.

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Effectively leveraging large, previously collected datasets in reinforcement learning (RL) is a key challenge for large-scale real-world applications. Offline RL algorithms promise to learn effective policies from previously-collected, static datasets without further interaction. However, in practice, offline RL presents a major challenge, and standard off-policy RL methods can fail due to overestimation of values induced by the distributional shift between the dataset and the learned policy, especially when training on complex and multi-modal data distributions. In this paper, we propose conservative Q-learning (CQL), which aims to address these limitations by learning a conservative Q-function such that the expected value of a policy under this Q-function lower-bounds its true value. We theoretically show that CQL produces a lower bound on the value of the current policy and that it can be incorporated into a policy learning procedure with theoretical improvement guarantees. In practice, CQL augments the standard Bellman error objective with a simple Q-value regularizer which is straightforward to implement on top of existing deep Q-learning and actor-critic implementations. On both discrete and continuous control domains, we show that CQL substantially outperforms existing offline RL methods, often learning policies that attain 2-5 times higher final return, especially when learning from complex and multi-modal data distributions. Preprint. Under review.

For agents

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

get_harvested_code_for_paper("2006.04779")
get_code_for_paper("2006.04779")
have("2006.04779")

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

curl https://syntology.ai/api/ran/2006.04779.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.04779.svg)](https://syntology.ai/paper/2006.04779)

Connect an agent — have() is free.