SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.08421 · 2024

Conservative and Risk-Aware Offline Multi-Agent Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 0 failures of our instrument; 1 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/2402.08421.json · badge: /api/ran/2402.08421.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
eslam211/conservative-and-distributional-marl canonical (classifier label)
classifier basis: census
2 ran · 1 no recorded run
FunctionStatusWhere it lives
eval_runs Ran
ran, honoured the contract we drafted
eslam211/conservative-and-distributional-marl/utils.py
code served (permissive licence) · get_code("40594cb711c4528b")
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
import torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleAgent(nn.Module):
    def __init__(self, state_dim=4, action_dim=3):
        super().__init__()
        self.net = nn.Linear(state_dim, action_dim)
    
    def get_action(self, state, epsilon):
        # state is a numpy array or list
        state_tensor = torch.FloatTensor(state).unsqueeze(0)
        with torch.no_grad():
            action_logits = self.net(state_tensor)
        action = action_logits.argmax(dim=1).numpy()
        return action

class SimpleEnv:
    def __init__(self, state_dim=4, num_agents=2):
        self.U = num_agents  # number of agents
        self.state_dim = state_dim
        self.step_count = 0
        self.max_steps = 5
        self.DONE = False
        self.Total_reward = 1.0
    
    def reset(self):
        self.step_count = 0
        self.DONE = False
        return np.random.randn(self.state_dim).tolist()
    
    def step(self, state, action):
        # state: list or array
        # action: list of integers (one per agent)
        self.step_count += 1
        
        # Generate next state
        next_state = np.random.randn(self.state_dim).tolist()
        
        # Assign reward
        reward = 1.0
        self.Total_reward = reward
        
        # Check if done
        done = self.step_count >= self.max_steps
        self.DONE = done
        
        return next_state, reward, done

# Create environment and agents
env = SimpleEnv(state_dim=4, num_agents=2)
agent = [SimpleAgent(state_dim=4, action_dim=3) for _ in range(env.U)]

ARGS = {
    "env": env,
    "agent": agent,
    "eval_runs": 5
}
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 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
eval_runs_dist Ran
ran, honoured the contract we drafted
eslam211/conservative-and-distributional-marl/utils.py
code served (permissive licence) · get_code("484f5e3d2b57fdbe")
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
import torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleAgent(nn.Module):
    def __init__(self, state_dim=4, action_dim=3):
        super().__init__()
        self.fc = nn.Linear(state_dim, action_dim)
    
    def get_action(self, state, epsilon):
        # Returns action (can be array or scalar)
        state_tensor = torch.tensor(state, dtype=torch.float32).unsqueeze(0)
        with torch.no_grad():
            action = self.fc(state_tensor).argmax(dim=1)
        return action.item()

class SimpleEnv:
    def __init__(self, num_agents=2, state_dim=4):
        self.U = num_agents  # number of agents
        self.state_dim = state_dim
        self.step_count = 0
        self.max_steps = 5
        self.DONE = False
        self.Total_reward = 1.0
    
    def reset(self):
        self.step_count = 0
        self.DONE = False
        return np.random.randn(self.state_dim)
    
    def step(self, state, action):
        self.step_count += 1
        next_state = np.random.randn(self.state_dim)
        reward = 1.0
        self.Total_reward = reward
        
        if self.step_count >= self.max_steps:
            self.DONE = True
        else:
            self.DONE = False
        
        done = self.DONE
        return next_state, reward, done

# Create instances
env = SimpleEnv(num_agents=2, state_dim=4)
agent = [SimpleAgent(state_dim=4, action_dim=3) for _ in range(2)]

ARGS = {
    'env': env,
    'agent': agent,
    'eval_runs': 5
}
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 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
prep_dataloader Not yet run
unverified (harvested, no recorded run)
eslam211/conservative-and-distributional-marl/utils.py
code served (permissive licence) · get_code("4ead42d8a31e115f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 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.

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

Reinforcement learning (RL) has been widely adopted for controlling and optimizing complex engineering systems such as next-generation wireless networks. An important challenge in adopting RL is the need for direct access to the physical environment. This limitation is particularly severe in multi-agent systems, for which conventional multi-agent reinforcement learning (MARL) requires a large number of coordinated online interactions with the environment during training. When only offline data is available, a direct application of online MARL schemes would generally fail due to the epistemic uncertainty entailed by the lack of exploration during training. In this work, we propose an offline MARL scheme that integrates distributional RL and conservative Q-learning to address the environment's inherent aleatoric uncertainty and the epistemic uncertainty arising from the use of offline data. We explore both independent and joint learning strategies. The proposed MARL scheme, referred to as multi-agent conservative quantile regression, addresses general risk-sensitive design criteria and is applied to the trajectory planning problem in drone networks, showcasing its advantages.

For agents

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

get_harvested_code_for_paper("2402.08421")
get_code_for_paper("2402.08421")
have("2402.08421")

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

curl https://syntology.ai/api/ran/2402.08421.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.08421.svg)](https://syntology.ai/paper/2402.08421)

Connect an agent — have() is free.