SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.17827 · ICLR · 2025

Langevin Soft Actor-Critic: Efficient Exploration through Uncertainty-Driven Critic Learning

Haque Ishfaq, Guangyuan Wang, Nur Sami, Doina Islam, Precup

arXiv · PDF · Open in the Atlas

Code that ran

6 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 3 failures of our instrument; 5 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/2501.17827.json · badge: /api/ran/2501.17827.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
hmishfaq/lsac canonical (classifier label)
classifier basis: classifier_2025
6 ran · 5 no recorded run
FunctionStatusWhere it lives
aSGLD Ran
ran on a synthesized input · constructed an object; no result was computed
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("d8b8cdf71799a5fb")
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 an instance of its own class, aSGLD: the run constructed the object and did not call it, so no result was computed. 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 torch.optim import Adam

torch.manual_seed(0)

# Create a simple model with parameters
model = nn.Sequential(
    nn.Linear(4, 3),
    nn.ReLU(),
    nn.Linear(3, 2)
)

# Create dummy data and compute a loss to generate gradients
x = torch.randn(2, 4)
y = torch.randn(2, 2)

# Forward pass to create gradients
output = model(x)
loss = nn.MSELoss()(output, y)
loss.backward()

# Now construct the ARGS dictionary for aSGLD
ARGS = {
    'params': model.parameters(),
    'lr': 0.001,
    'betas': (0.9, 0.999),
    'eps': 1e-8,
    'weight_decay': 0.0,
    'amsgrad': False,
    'noise_scale': 0.01,
    'a': 1.0
}
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 aSGLD — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a aSGLD 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
combined_shape Ran
ran, honoured the contract we drafted
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("6d24f19c56e16f4a")
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 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 numpy as np

ARGS = {
    "length": 5,
    "shape": None,
}
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 tuple, ndim 1, dtype family int — 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
create_apprfunc Ran
ran; our contract draft was wrong, not the code
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("40e25432a2f48697")
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 TestApprfunc. 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 unittest.mock import MagicMock
import sys

torch.manual_seed(0)

# Create a minimal mock module for 'networks.test_apprfunc'
class TestApprfunc(nn.Module):
    def __init__(self, **kwargs):
        super().__init__()
        self.kwargs = kwargs
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create mock network module
mock_network_module = MagicMock()
mock_network_module.TestApprfunc = TestApprfunc

# Inject the mock module into sys.modules so importlib can find it
sys.modules['networks'] = MagicMock()
sys.modules['networks.test_apprfunc'] = mock_network_module

ARGS = {
    'apprfunc': 'test_apprfunc',
    'name': 'test_apprfunc'
}
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 TestApprfunc — a TestApprfunc output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a TestApprfunc 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
formatter Ran
ran; our contract draft was wrong, not the code
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("5e059d3b24974ac1")
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 torch

def formatter(src: str, firstUpper: bool=True):
    arr = src.split('_')
    res = ''
    for i in arr:
        res = res + i[0].upper() + i[1:]
    if not firstUpper:
        res = res[0].lower() + res[1:]
    return res

ARGS = {
    'src': 'hello_world_example',
    'firstUpper': True
}
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 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
lr_lambda Ran
ran, honoured the contract we drafted
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("e570f37628252c79")
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 float, 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

torch.manual_seed(0)

# Analyze the function signature and body:
# - it: an integer representing the current iteration
# - start_iter: an integer with default value 100000
# The function uses these parameters in arithmetic operations to compute a learning rate multiplier

ARGS = {
    "it": 150000,
    "start_iter": 100000,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
set_seed Ran
ran; our contract draft was wrong, not the code
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("568a3d72b42e3f4b")
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
import numpy as np

torch.manual_seed(0)

class StandInEnv:
    """Stand-in environment with a seed method."""
    def seed(self, seed_value):
        np.random.seed(seed_value)
        return [seed_value]

trainer_name = "test_trainer"
seed = 42
offset = 10
env = StandInEnv()

ARGS = {
    "trainer_name": trainer_name,
    "seed": seed,
    "offset": offset,
    "env": env
}
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 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
DiffusionBuffer Not yet run
unverified (harvested, no recorded run)
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("fcf8c24349da40d7")
How it ran

no invocation recorded.

LSAC Not yet run
unverified (harvested, no recorded run)
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("af857be3d6b0ff73")
How it ran

no invocation recorded.

Networks Not yet run
unverified (harvested, no recorded run)
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("777af9a7506f47f8")
How it ran

no invocation recorded.

adam_sgld Not yet run
unverified (harvested, no recorded run)
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("a80ed6021e69c6b7")
How it ran

no invocation recorded.

get_apprfunc_dict Not yet run
unverified (harvested, no recorded run)
hmishfaq/lsac/lsac.py
code served (permissive licence) · get_code("6f67468e83cdf046")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 11 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.

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

Existing actor-critic algorithms, which are popular for continuous control reinforcement learning (RL) tasks, suffer from poor sample efficiency due to lack of principled exploration mechanism within them. Motivated by the success of Thompson sampling for efficient exploration in RL, we propose a novel model-free RL algorithm, Langevin Soft Actor Critic (LSAC), which prioritizes enhancing critic learning through uncertainty estimation over policy optimization. LSAC employs three key innovations: approximate Thompson sampling through distributional Langevin Monte Carlo (LMC) based Q updates, parallel tempering for exploring multiple modes of the posterior of the Q function, and diffusion synthesized state-action samples regularized with Q action gradients. Our extensive experiments demonstrate that LSAC outperforms or matches the performance of mainstream model-free RL algorithms for continuous control tasks. Notably, LSAC marks the first successful application of an LMC based Thompson sampling in continuous control tasks with continuous action spaces.

For agents

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

get_harvested_code_for_paper("2501.17827")
get_code_for_paper("2501.17827")
have("2501.17827")

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

curl https://syntology.ai/api/ran/2501.17827.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.17827.svg)](https://syntology.ai/paper/2501.17827)

Connect an agent — have() is free.