SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2110.02034 · ICLR · 2022

Dropout Q-Functions for Doubly Efficient Reinforcement Learning

Taisei Hashimoto, Yoshimasa Tsuruoka, Takuya Hiraoka, Takahisa Imagawa, Takashi Onishi

arXiv · PDF · Open in the Atlas

Code that ran

6 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 1 failure of our instrument; 2 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/2110.02034.json · badge: /api/ran/2110.02034.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
takuyahiraoka/dropout-q-functions-for-doubly-efficient-reinforcement-learning canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
ku2482/soft-actor-critic.pytorch canonical (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
watchernyu/REDQ community
Papers with Code archive
4 ran · 2 no recorded run
FunctionStatusWhere it lives
Mlp Ran
ran on a synthesized input · constructed an object; no result was computed
watchernyu/REDQ/redq/algos/redq_sac.py
code served (permissive licence) · get_code("3ec1489c1ccfeec3")
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, Mlp: 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.nn import functional as F

torch.manual_seed(0)

ARGS = {
    "input_size": 4,
    "output_size": 3,
    "hidden_sizes": [8, 6],
    "hidden_activation": F.relu,
}
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 Mlp — 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 Mlp 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
ReplayBuffer Ran
ran on a synthesized input · constructed an object; no result was computed
watchernyu/REDQ/redq/algos/redq_sac.py
code served (permissive licence) · get_code("820da6f0b914ce98")
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, ReplayBuffer: 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 numpy as np

ARGS = {
    "obs_dim": 4,
    "act_dim": 3,
    "size": 16,
}
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 ReplayBuffer — 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 ReplayBuffer 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
TanhGaussianPolicy Ran
ran on a synthesized input · constructed an object; no result was computed
watchernyu/REDQ/redq/algos/redq_sac.py
code served (permissive licence) · get_code("5fb36a006ce8ed7c")
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, TanhGaussianPolicy: 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
import torch.nn.functional as F

torch.manual_seed(0)

ARGS = {
    "obs_dim": 4,
    "action_dim": 3,
    "hidden_sizes": [8, 8],
    "hidden_activation": F.relu,
    "action_limit": 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 TanhGaussianPolicy — 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 TanhGaussianPolicy 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
get_probabilistic_num_min Ran
ran, honoured the contract we drafted
watchernyu/REDQ/redq/algos/redq_sac.py
code served (permissive licence) · get_code("dfde273b47c42786")
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 int, ndim 0, dtype family int, 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 numpy as np

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

# The function takes a single parameter: num_mins
# It should be a float or numeric value that represents a number of minutes
# The function floors it and uses probabilistic rounding based on the fractional part

ARGS = {
    "num_mins": 3.7
}
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 int, ndim 0, dtype family int, 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
test_agent Ran
ran, honoured the contract we drafted
takuyahiraoka/dropout-q-functions-for-doubly-efficient-reinforcement-learning/OriginalREDQCodebase/redq/algos/core.py
code served (permissive licence) · get_code("88623398df64361e")
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 (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch
import torch.nn as nn
from torch.nn import functional as F
from torch.distributions import Distribution, Normal

torch.manual_seed(0)

# Create a minimal agent with get_test_action method
class MinimalAgent(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def get_test_action(self, observation):
        """Return a deterministic action given observation"""
        obs_tensor = torch.tensor(observation, dtype=torch.float32).unsqueeze(0)
        action = self.linear(obs_tensor).squeeze(0).detach().numpy()
        return action

# Create a minimal environment
class MinimalEnv:
    def __init__(self):
        self.obs_dim = 4
        self.action_dim = 3
        self.step_count = 0
    
    def reset(self):
        """Reset environment and return initial observation"""
        self.step_count = 0
        return np.random.randn(self.obs_dim).astype(np.float32)
    
    def step(self, action):
        """Execute one step in environment"""
        self.step_count += 1
        observation = np.random.randn(self.obs_dim).astype(np.float32)
        reward = float(np.random.randn())
        done = self.step_count >= 5  # Episode ends after 5 steps
        info = {}
        return observation, reward, done, info

# Create a minimal logger
class MinimalLogger:
    def __init__(self):
        self.data = {}
    
    def store(self, **kwargs):
        """Store logged values"""
        for key, value in kwargs.items():
            if key not in self.data:
                self.data[key] = []
            self.data[key].append(value)

# Construct arguments
agent = MinimalAgent()
test_env = MinimalEnv()
max_ep_len = 10
logger = MinimalLogger()
n_eval = 2

ARGS = {
    "agent": agent,
    "test_env": test_env,
    "max_ep_len": max_ep_len,
    "logger": logger,
    "n_eval": n_eval
}
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 ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
to_batch Ran
ran; our contract draft was wrong, not the code
ku2482/soft-actor-critic.pytorch/code/utils.py
code served (permissive licence) · get_code("3afd917c65b80e08")
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. 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)

ARGS = {
    "state": [1.0, 2.0, 3.0, 4.0],
    "action": 0.5,
    "reward": 1.0,
    "next_state": [2.0, 3.0, 4.0, 5.0],
    "done": 1.0,
    "device": torch.device("cpu")
}
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 — a tuple output has no array shape of its own; none is 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
REDQSACAgent Not yet run
unverified (harvested, no recorded run)
watchernyu/REDQ/redq/algos/redq_sac.py
code served (permissive licence) · get_code("c122b3cd8b4c6346")
How it ran

no invocation recorded.

soft_update_model1_with_model2 Not yet run
unverified (harvested, no recorded run)
watchernyu/REDQ/redq/algos/redq_sac.py
code served (permissive licence) · get_code("f95d746979ff7da6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 2 later) and 3 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
Mlp 2 other papers
2 dated earlier · 2 only as an assembly brick
1801.01290 (2018-01)
Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
2101.05982 (2021-01)
Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
soft_update_model1_with_model2 2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
2101.05982 (2021-01)
Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
2605.05863 (2026-05)
SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior Data
to_batch 2 other papers
1 dated earlier, 1 later
1910.07207 (2019-10)
Soft Actor-Critic for Discrete Action Settings
2605.21282 (2026-05)
Stochastic MeanFlow Policies: One-Step Generative Control with Entropic Mirror Descent
ReplayBuffer 1 other paper
1 dated earlier · 1 only as an assembly brick
2101.05982 (2021-01)
Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
get_probabilistic_num_min 1 other paper
1 dated earlier · 1 only as an assembly brick
2101.05982 (2021-01)
Randomized Ensembled Double Q-Learning: Learning Fast Without a Model

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

Randomized ensembled double Q-learning (REDQ) (Chen et al., 2021b) has recently achieved state-of-the-art sample efficiency on continuous-action reinforcement learning benchmarks. This superior sample efficiency is made possible by using a large Q-function ensemble. However, REDQ is much less computationally efficient than non-ensemble counterparts such as Soft Actor-Critic (SAC) (Haarnoja et al., 2018a). To make REDQ more computationally efficient, we propose a method of improving computational efficiency called DroQ, which is a variant of REDQ that uses a small ensemble of dropout Q-functions. Our dropout Q-functions are simple Q-functions equipped with dropout connection and layer normalization. Despite its simplicity of implementation, our experimental results indicate that DroQ is doubly (sample and computationally) efficient. It achieved comparable sample efficiency with REDQ, much better computational efficiency than REDQ, and comparable computational efficiency with that of SAC.

For agents

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

get_harvested_code_for_paper("2110.02034")
get_code_for_paper("2110.02034")
have("2110.02034")

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

curl https://syntology.ai/api/ran/2110.02034.json

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

[![Syntology run record](https://syntology.ai/api/ran/2110.02034.svg)](https://syntology.ai/paper/2110.02034)

Connect an agent — have() is free.