SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2001.03415 · ICLR · 2020

Multi-Agent Interactions Modeling with Correlated Policies

Minghuan Liu, Ming Zhou, Weinan Zhang, Yuzheng Zhuang, Jun Wang, Wulong Liu, Yong Yu

arXiv · PDF · Open in the Atlas

Code that ran

7 of 13 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; 5 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 1 failure of our instrument; 6 no recorded run), measured 2026-11-21; 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/2001.03415.json · badge: /api/ran/2001.03415.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
apexrl/codail canonical (classifier label)
classifier basis: classifier_2025
0 ran · 3 no recorded run
apexrl/CoDAIL marked official in Papers with Code
Papers with Code archive
7 ran · 3 no recorded run
FunctionStatusWhere it lives
Scheduler Ran
ran on a synthesized input · constructed an object; no result was computed
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("dee1879477183342")
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, Scheduler: 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 numpy as np

torch.manual_seed(0)

# Examining the Scheduler class:
# __init__ takes: v (numeric), nvalues (numeric), schedule (string key)
# value() method uses: self.v, self.schedule, self.n, self.nvalues
# value_steps() method uses: self.v, self.schedule, self.nvalues
# schedule parameter must be a key in the 'schedules' dict

ARGS = {
    'v': 0.5,
    'nvalues': 100,
    'schedule': 'linear'
}
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 Scheduler — 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 Scheduler 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
constant Ran
ran, honoured the contract we drafted
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("51b561109c463fd3")
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 torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "p": 0.5
}
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 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
double_linear_con Ran
ran, honoured the contract we drafted
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("cda9ff3c49392a4d")
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 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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - Parameter 'p' is used with *= 2 (multiplication assignment)
# - Then compared with 1 - p < eps
# - Returns either eps or 1 - p
# This suggests p should be a numeric value, likely a float between 0 and 1

p = 0.5

ARGS = {
    "p": p
}
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 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
double_middle_drop Ran
ran, honoured the contract we drafted
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("7675c23be44244ab")
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 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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - Takes parameter 'p'
# - Performs numeric comparisons and arithmetic on p
# - Returns a numeric value
# - No complex object interactions needed

# The function expects p to be a numeric value that can be compared and used in arithmetic
# We'll test with a value that exercises the logic paths

p = 0.5

ARGS = {
    "p": p
}
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 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
linear Ran
ran, honoured the contract we drafted
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("edab60a37c16451e")
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 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)

ARGS = {
    "p": 0.5
}
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 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
middle_drop Ran
ran, honoured the contract we drafted
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("aea387f414294fd5")
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 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)

# The function middle_drop takes a single parameter 'p' which is used in numeric comparisons
# It appears to be a dropout probability value (float between 0 and 1)

ARGS = {
    "p": 0.5
}
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 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
ortho_init Ran
ran; our contract draft was wrong, not the code
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("0cb9b8299fe38738")
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 function. 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

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

# The ortho_init function takes a scale parameter
# The returned _ortho_init function takes shape, dtype, and partition_info
# We need to test the ortho_init function by calling it and then using the returned initializer

scale_value = 2.0

ARGS = {
    "scale": scale_value
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
Discriminator Not yet run
unverified (harvested, no recorded run)
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("61a8673dcf121b59")
How it ran

no invocation recorded.

fc Not yet run
unverified (harvested, no recorded run)
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("6a7b574670cffc86")
How it ran

no invocation recorded.

find_task_for_env_id_in_any_benchmark Not yet run
unverified (harvested, no recorded run)
apexrl/codail/multi-agent-irl/rl/bench/benchmarks.py
code served (permissive licence) · get_code("97c4da6e50f18d58")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

find_trainable_variables Not yet run
unverified (harvested, no recorded run)
apexrl/CoDAIL/multi-agent-irl/irl/mack/kfac_discriminator_codail.py
code served (permissive licence) · get_code("dc11f6f1b3db954a")
How it ran

no invocation recorded.

get_benchmark Not yet run
unverified (harvested, no recorded run)
apexrl/codail/multi-agent-irl/rl/bench/benchmarks.py
code served (permissive licence) · get_code("315ed50e400048ee")
How it ran

no invocation recorded.

get_task Not yet run
unverified (harvested, no recorded run)
apexrl/codail/multi-agent-irl/rl/bench/benchmarks.py
code served (permissive licence) · get_code("7e1a88346f1d7ae7")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 4 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.

FunctionAlso present inEarliest-datedOthers named
find_task_for_env_id_in_any_benchmark 5 other papers
3 dated earlier, 2 later
1604.06778 (2016-04)
Benchmarking Deep Reinforcement Learning for Continuous Control
1606.03476 (2016-06)
Generative Adversarial Imitation Learning
1905.02363 (2019-05)
Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning
and 2 more not named here
get_benchmark 5 other papers
3 dated earlier, 2 later
1604.06778 (2016-04)
Benchmarking Deep Reinforcement Learning for Continuous Control
1606.03476 (2016-06)
Generative Adversarial Imitation Learning
1905.02363 (2019-05)
Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning
and 2 more not named here
get_task 5 other papers
3 dated earlier, 2 later
1604.06778 (2016-04)
Benchmarking Deep Reinforcement Learning for Continuous Control
1606.03476 (2016-06)
Generative Adversarial Imitation Learning
1905.02363 (2019-05)
Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning
and 2 more not named here
ortho_init
brick of this paper's assembly: helper
4 other papers
4 later · 2 only as an assembly brick
2010.07916 (2020-10)
Multi-Agent Trust Region Policy Optimization
2103.12300 (2021-03)
Drop-Bottleneck: LEARNING DISCRETE COMPRESSED REPRESENTATION FOR NOISE-ROBUST EXPLORATION
2105.09452 (2021-05)
Minimum-Delay Adaptation in Non-Stationary Reinforcement Learning via Online High-Confidence Change-Point Detection
and 1 more not named here

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

In multi-agent systems, complex interacting behaviors arise due to the high correlations among agents. However, previous work on modeling multi-agent interactions from demonstrations is primarily constrained by assuming the independence among policies and their reward structures. In this paper, we cast the multiagent interactions modeling problem into a multi-agent imitation learning framework with explicit modeling of correlated policies by approximating opponents' policies, which can recover agents' policies that can regenerate similar interactions. Consequently, we develop a Decentralized Adversarial Imitation Learning algorithm with Correlated policies (CoDAIL), which allows for decentralized training and execution. Various experiments demonstrate that CoDAIL can better regenerate complex interactions close to the demonstrators and outperforms state-of-the-art multi-agent imitation learning methods. Our code is available at https://github.com/apexrl/CoDAIL.

For agents

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

get_harvested_code_for_paper("2001.03415")
get_code_for_paper("2001.03415")
have("2001.03415")

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

curl https://syntology.ai/api/ran/2001.03415.json

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

[![Syntology run record](https://syntology.ai/api/ran/2001.03415.svg)](https://syntology.ai/paper/2001.03415)

Connect an agent — have() is free.