SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1908.09453 · 2019

OpenSpiel: A Framework for Reinforcement Learning in Games

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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; 1 honoured the contract we drafted, 1 violated it and 6 have no contract verdict; 0 failures of our instrument; 0 no recorded run), 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/1908.09453.json · badge: /api/ran/1908.09453.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
wyz2368/open_spiel_egta community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
Tyjch/open_spiel community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
qmaai/open_spiel community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
exhaustive_strategy_selector Ran
ran on a synthesized input
wyz2368/open_spiel_egta/open_spiel/python/algorithms/psro_variations/generalized_egta.py
code served (permissive licence) · get_code("c5983134be459f12")
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 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 torch.nn as nn

torch.manual_seed(0)

class MockGenPSROSolver:
    """Mock GenPSROSolver with get_policies method."""
    
    def __init__(self):
        self._policies = [
            {"player_0": "policy_a", "player_1": "policy_x"},
            {"player_0": "policy_b", "player_1": "policy_y"},
        ]
    
    @property
    def get_policies(self):
        """Returns the policies for all players."""
        return self._policies


solver = MockGenPSROSolver()

ARGS = {
    "solver": solver
}
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 list, ndim 1, dtype family float — 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
lens Ran
ran, honoured the contract we drafted
Tyjch/open_spiel/open_spiel/python/algorithms/double_oracle.py
code served (permissive licence) · get_code("e893a592cd4dd8fe")
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

torch.manual_seed(0)

# The function lens() takes a parameter 'lists' which should be a list of lists
# and returns the length of each list. We need to provide a list of lists with
# different sizes to make the test meaningful.

ARGS = {
    "lists": [
        [1, 2, 3],
        [4, 5],
        [6, 7, 8, 9],
        []
    ]
}
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 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
probabilistic_strategy_selector Ran
ran on a synthesized input
wyz2368/open_spiel_egta/open_spiel/python/algorithms/psro_variations/generalized_egta.py
code served (permissive licence) · get_code("51e8afabe25ebd98")
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 2, 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

class MockGenPSROSolver:
    def __init__(self):
        self.get_kwargs = {
            'number_policies_selected': 2
        }
        self.get_policies = [
            ['policy_0_0', 'policy_0_1', 'policy_0_2'],
            ['policy_1_0', 'policy_1_1']
        ]
        self._meta_strategies = [
            np.array([0.5, 0.3, 0.2]),
            np.array([0.6, 0.4])
        ]
    
    def get_and_update_meta_strategies(self, update=False):
        return self._meta_strategies

solver = MockGenPSROSolver()

ARGS = {
    'solver': solver
}
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 list, ndim 2, dtype family float — 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
rectified_strategy_selector Ran
ran on a synthesized input
wyz2368/open_spiel_egta/open_spiel/python/algorithms/psro_variations/generalized_egta.py
code served (permissive licence) · get_code("c2404a4c6f480a8e")
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. 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)

class MockSolver:
    def __init__(self):
        self.policies_data = [
            ["policy_0_0", "policy_0_1", "policy_0_2"],
            ["policy_1_0", "policy_1_1"]
        ]
        self.meta_strategies_data = [
            np.array([0.5, 1e-7, 0.5]),
            np.array([0.3, 2e-7])
        ]
    
    @property
    def get_policies(self):
        return self.policies_data
    
    def get_and_update_meta_strategies(self, update=False):
        return self.meta_strategies_data

solver = MockSolver()

ARGS = {
    "solver": solver
}
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 — a list 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
renormalize Ran
ran on a synthesized input
wyz2368/open_spiel_egta/open_spiel/python/algorithms/psro_variations/abstract_meta_trainer_egta.py
code served (permissive licence) · get_code("8469ed1452bc3dd0")
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 (5,) 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

# Create a simple probability vector with some negative values and some positive values
# The function will set negatives to 0 and then normalize
probabilities = np.array([0.5, -0.2, 0.3, -0.1, 0.4], dtype=np.float32)

ARGS = {
    "probabilities": probabilities
}
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 1, dtype family float, shape (5,) — shape (5,) 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
sample_episode Ran
ran on a synthesized input
wyz2368/open_spiel_egta/open_spiel/python/algorithms/psro_variations/abstract_meta_trainer_egta.py
code served (permissive licence) · get_code("c0276e4601634e38")
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 (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

class MockState:
    """Mock Pyspiel state for testing."""
    def __init__(self, is_terminal=False, is_chance=False, current_player_val=0, depth=0):
        self.is_terminal_val = is_terminal
        self.is_chance_val = is_chance
        self.current_player_val = current_player_val
        self.depth = depth
        self.max_depth = 3
    
    def is_terminal(self):
        return self.is_terminal_val
    
    def is_chance_node(self):
        return self.is_chance_val
    
    def current_player(self):
        return self.current_player_val
    
    def chance_outcomes(self):
        """Returns list of (action, probability) tuples for chance nodes."""
        return [(0, 0.5), (1, 0.5)]
    
    def apply_action(self, action):
        """Simulate applying an action."""
        self.depth += 1
        if self.depth >= self.max_depth:
            self.is_terminal_val = True
            self.is_chance_val = False
        else:
            self.is_chance_val = not self.is_chance_val
            if not self.is_chance_val:
                self.current_player_val = 1 - self.current_player_val
    
    def returns(self):
        """Return final payoffs."""
        return [1.0, -1.0]


def create_policy_0(state):
    """Policy for player 0: returns action distribution."""
    return {0: 0.3, 1: 0.7}


def create_policy_1(state):
    """Policy for player 1: returns action distribution."""
    return {0: 0.6, 1: 0.4}


ARGS = {
    "state": MockState(is_terminal=False, is_chance=True, current_player_val=0),
    "policies": [create_policy_0, create_policy_1]
}
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 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
softmax Ran
ran, violated the contract we drafted
qmaai/open_spiel/open_spiel/python/algorithms/ars.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from thuml/DARE, is pointer only (licence: NONE) · get_code("a585b44308980803")
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 thuml/DARE, 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.

uniform_strategy Ran
ran on a synthesized input
wyz2368/open_spiel_egta/open_spiel/python/algorithms/psro_variations/abstract_meta_trainer_egta.py
code served (permissive licence) · get_code("8faf6f7ec3ffb422")
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 2, 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

class MockGenPSROSolver:
    """Mock solver that implements the interface expected by uniform_strategy."""
    
    def __init__(self, policies):
        self._policies = policies
    
    @property
    def get_policies(self):
        return self._policies


# Test case 1: Single policy (list of scalars)
solver_single = MockGenPSROSolver([0.5, 0.3, 0.2])

# Test case 2: Multiple policies (list of lists)
solver_multiple = MockGenPSROSolver([[0.4, 0.6], [0.25, 0.25, 0.5]])

# Use the single policy solver as the main test case
ARGS = {
    "solver": solver_single
}
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 list, ndim 2, dtype family float — 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

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, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 6 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
softmax 3 other papers
3 later
2010.12777 (2020-10)
Improving Multilingual Models with Language-Clustered Vocabularies
2104.00795 (2021-04)
No Cost Likelihood Manipulation at Test Time for Making Better Mistakes in Deep Networks
2410.02604 (2024-10)
Long-Sequence Recommendation Models Need Decoupled Embeddings
lens 2 other papers
2 later
2402.19420 (2024-02)
Understanding Iterative Combinatorial Auction Designs via Multi-Agent Reinforcement Learning
2602.05591 (2026-02)
Efficient Algorithms for Robust Markov Decision Processes with s-Rectangular Ambiguity Sets

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

OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games. OpenSpiel supports n-player (single- and multi- agent) zero-sum, cooperative and general-sum, one-shot and sequential, strictly turn-taking and simultaneous-move, perfect and imperfect information games, as well as traditional multiagent environments such as (partially- and fully- observable) grid worlds and social dilemmas. OpenSpiel also includes tools to analyze learning dynamics and other common evaluation metrics. This document serves both as an overview of the code base and an introduction to the terminology, core concepts, and algorithms across the fields of reinforcement learning, computational game theory, and search.

For agents

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

get_harvested_code_for_paper("1908.09453")
get_code_for_paper("1908.09453")
have("1908.09453")

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

curl https://syntology.ai/api/ran/1908.09453.json

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

[![Syntology run record](https://syntology.ai/api/ran/1908.09453.svg)](https://syntology.ai/paper/1908.09453)

Connect an agent — have() is free.