SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2112.07013 · 2021

PantheonRL: A MARL Library for Dynamic Training Interactions

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2112.07013.json · badge: /api/ran/2112.07013.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
Stanford-ILIAD/PantheonRL canonical (classifier label)
classifier basis not recorded on the link
9 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_space Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/util.py
code served (permissive licence) · get_code("5742ccbbeea55d95")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Box. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from gym.spaces import Box, Discrete, MultiBinary, MultiDiscrete

# Test with a Box space
box_space = Box(low=np.array([0.0, 0.0]), high=np.array([1.0, 1.0]), dtype=np.float32)
numframes = 3

ARGS = {
    "space": box_space,
    "numframes": numframes
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Box — a Box output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Box output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
dataclass_quick_asdict Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/trajsaver.py
code served (permissive licence) · get_code("e7e69361998f84c0")
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 dataclasses
import torch
import numpy as np

torch.manual_seed(0)

@dataclasses.dataclass
class SampleDataclass:
    """A simple dataclass with various field types for testing."""
    name: str
    values: np.ndarray
    tensor_data: torch.Tensor
    count: int
    ratio: float

# Create an instance of the dataclass
instance = SampleDataclass(
    name="test_sample",
    values=np.array([1.0, 2.0, 3.0, 4.0]),
    tensor_data=torch.randn(2, 4),
    count=42,
    ratio=0.75
)

ARGS = {
    "dataclass_instance": instance
}
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
extract_obs Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/observation.py
code served (permissive licence) · get_code("1005123632875338")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from dataclasses import dataclass
from typing import Optional

@dataclass
class Observation:
    """
    Representation of a single observation provided by an environment.

    obs: the (partial) observation that an agent receives
    state: the full state information
    action_mask: a mask specifying what actions are legal
        - if it is None, all actions are permitted
    """
    obs: np.ndarray
    state: np.ndarray
    action_mask: Optional[np.ndarray] = None

    def __init__(self, obs: np.ndarray, state: Optional[np.ndarray]=None, action_mask: Optional[np.ndarray]=None):
        self.obs = obs
        self.state = state if state is not None else obs
        self.action_mask = action_mask

# Create sample observation data
obs_array = np.random.randn(4, 3).astype(np.float32)
state_array = np.random.randn(4, 5).astype(np.float32)
action_mask = np.array([1, 0, 1, 1], dtype=np.int32)

# Create an Observation instance
observation = Observation(
    obs=obs_array,
    state=state_array,
    action_mask=action_mask
)

ARGS = {
    "observation": observation
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 3) — shape (4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
extract_partial_obs Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/observation.py
code served (permissive licence) · get_code("ecf30766f5776b7c")
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 numpy as np
from dataclasses import dataclass
from typing import Optional

@dataclass
class Observation:
    """
    Representation of a single observation provided by an environment.

    obs: the (partial) observation that an agent receives
    state: the full state information
    action_mask: a mask specifying what actions are legal
        - if it is None, all actions are permitted
    """
    obs: np.ndarray
    state: np.ndarray
    action_mask: Optional[np.ndarray] = None

    def __init__(self, obs: np.ndarray, state: Optional[np.ndarray]=None, action_mask: Optional[np.ndarray]=None):
        self.obs = obs
        self.state = state if state is not None else obs
        self.action_mask = action_mask

obs_array = np.random.randn(4, 3).astype(np.float32)
state_array = np.random.randn(4, 5).astype(np.float32)
action_mask_array = np.array([1, 1, 0, 1], dtype=np.float32)

observation = Observation(
    obs=obs_array,
    state=state_array,
    action_mask=action_mask_array
)

ARGS = {
    "observation": observation
}
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 — 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
get_L2_sphere Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/algos/adap/util.py
code served (permissive licence) · get_code("ef33952ecde12340")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch as th
import numpy as np

th.manual_seed(0)

ARGS = {
    "ctx_size": 4,
    "num": 2,
    "torch": False
}
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 ndarray, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_default_obs Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/util.py
code served (permissive licence) · get_code("df6a9974418bb1ed")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (4,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import gym
from gym.spaces import Box, Discrete, MultiBinary, MultiDiscrete
import numpy as np

class SimpleEnv(gym.Env):
    """Minimal environment for testing"""
    def __init__(self, space_type='box'):
        super().__init__()
        self.space_type = space_type
        
        if space_type == 'box':
            self.observation_space = Box(low=np.array([0.0, -1.0, -2.0, -3.0]), 
                                        high=np.array([1.0, 2.0, 3.0, 4.0]), 
                                        dtype=np.float32)
        elif space_type == 'discrete':
            self.observation_space = Discrete(5)
        elif space_type == 'multibinary':
            self.observation_space = MultiBinary(4)
        elif space_type == 'multidiscrete':
            self.observation_space = MultiDiscrete([3, 4, 5, 2])
        
        self.action_space = Discrete(2)
    
    def reset(self):
        return self.observation_space.sample()
    
    def step(self, action):
        obs = self.observation_space.sample()
        reward = 0.0
        done = False
        info = {}
        return obs, reward, done, info

env_box = SimpleEnv(space_type='box')
env_discrete = SimpleEnv(space_type='discrete')
env_multibinary = SimpleEnv(space_type='multibinary')
env_multidiscrete = SimpleEnv(space_type='multidiscrete')

ARGS = {
    'env': env_box
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (4,) — shape (4,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_space_size Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/util.py
code served (permissive licence) · get_code("fb6ab3a1f1defb2a")
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
from gym.spaces import Box, Discrete, MultiBinary, MultiDiscrete

# Test with different space types
box_space = Box(low=np.array([0.0, 0.0, 0.0, 0.0]), high=np.array([1.0, 1.0, 1.0, 1.0]))
discrete_space = Discrete(5)
multibinary_space = MultiBinary(3)
multidiscrete_space = MultiDiscrete([2, 3, 4])

ARGS = {
    "space": box_space
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
get_unit_square Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/algos/adap/util.py
code served (permissive licence) · get_code("865642e814e5a593")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch as th
import numpy as np
th.manual_seed(0)

ARGS = {
    "ctx_size": 4,
    "num": 2,
    "torch": False
}
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 ndarray, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
transitions_collate_fn Ran
ran on a synthesized input
Stanford-ILIAD/PantheonRL/pantheonrl/common/trajsaver.py
code served (permissive licence) · get_code("a3a303088488f5f0")
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
from typing import Dict, Mapping, Sequence, Union

torch.manual_seed(0)

# Create a batch of transition samples
# Each sample is a mapping with string keys and numpy array values
sample1 = {
    "obs": np.array([[1.0, 2.0, 3.0, 4.0]], dtype=np.float32),
    "action": np.array([[0.5, 0.6]], dtype=np.float32),
    "next_obs": np.array([[1.1, 2.1, 3.1, 4.1]], dtype=np.float32),
    "reward": np.array([1.0], dtype=np.float32),
    "done": np.array([False], dtype=bool),
}

sample2 = {
    "obs": np.array([[5.0, 6.0, 7.0, 8.0]], dtype=np.float32),
    "action": np.array([[0.7, 0.8]], dtype=np.float32),
    "next_obs": np.array([[5.1, 6.1, 7.1, 8.1]], dtype=np.float32),
    "reward": np.array([2.0], dtype=np.float32),
    "done": np.array([True], dtype=bool),
}

batch = [sample1, sample2]

ARGS = {
    "batch": batch,
}
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

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 9 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

We present PantheonRL, a multiagent reinforcement learning software package for dynamic training interactions such as round-robin, adaptive, and ad-hoc training. Our package is designed around flexible agent objects that can be easily configured to support different training interactions, and handles fully general multiagent environments with mixed rewards and n agents. Built on top of StableBaselines3, our package works directly with existing powerful deep RL algorithms. Finally, PantheonRL comes with an intuitive yet functional web user interface for configuring experiments and launching multiple asynchronous jobs. Our package can be found at https://github.com/Stanford-ILIAD/PantheonRL.

For agents

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

get_harvested_code_for_paper("2112.07013")
get_code_for_paper("2112.07013")
have("2112.07013")

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

curl https://syntology.ai/api/ran/2112.07013.json

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

[![Syntology run record](https://syntology.ai/api/ran/2112.07013.svg)](https://syntology.ai/paper/2112.07013)

Connect an agent — have() is free.