SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2206.09670 · 2022

Benchmarking Constraint Inference in Inverse Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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 2 have no contract verdict; 2 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/2206.09670.json · badge: /api/ran/2206.09670.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
guiliang/cirl-benchmarks-public canonical (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
guiliang/icrl-benchmarks-public canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 1 no recorded run
FunctionStatusWhere it lives
get_net_arch Ran
ran on a synthesized input
guiliang/cirl-benchmarks-public/utils/model_utils.py
code served (permissive licence) · get_code("4836f9d7b886fc29")
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 io

# Create a small config dictionary with the required PPO structure
config = {
    'PPO': {
        'policy_layers': [64, 32],
        'reward_vf_layers': [64, 32],
        'cost_vf_layers': [64, 32]
    }
}

# Create a log file (in-memory StringIO to avoid actual file I/O)
log_file = io.StringIO()

ARGS = {
    'config': config,
    'log_file': log_file
}
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
get_plot_results_dir Ran
ran; our contract draft was wrong, not the code
guiliang/icrl-benchmarks-public/interface/plot_results/plot_results_dirs.py
code served (permissive licence) · get_code("f08b27214ee31ece")
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 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.

# Analyzing the function, it takes env_id as a parameter and returns a dictionary
# based on the env_id value. The function doesn't actually use the returned value
# in any complex way - it just constructs and returns a dict of log paths.

# The function signature is:
# def get_plot_results_dir(env_id):

# The parameter env_id is a string that determines which log_path_dict to return.
# Looking at the function body, it checks if env_id == 'highD_velocity_constraint'
# and returns a specific dictionary. We need to provide a valid env_id string.

ARGS = {
    'env_id': 'highD_velocity_constraint'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — 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
handle_model_parameters Ran
ran on a synthesized input
guiliang/cirl-benchmarks-public/utils/model_utils.py
code served (permissive licence) · get_code("58e77c9c2ec538c6")
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
import torch.nn as nn
from io import StringIO

torch.manual_seed(0)

# Create a simple model with named parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.encoder_layer1 = nn.Linear(4, 3)
        self.encoder_layer2 = nn.Linear(3, 2)
        self.decoder_layer1 = nn.Linear(2, 3)
        self.other_layer = nn.Linear(3, 4)
    
    def forward(self, x):
        return x

model = SimpleModel()

# Keywords to identify active parameters
active_keywords = ['encoder']

# Model name for logging
model_name = 'TestModel'

# Log file
log_file = StringIO()

# Flag to set require_grad
set_require_grad = True

ARGS = {
    'model': model,
    'active_keywords': active_keywords,
    'model_name': model_name,
    'log_file': log_file,
    'set_require_grad': set_require_grad
}
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
masked_softmax Ran
ran; our contract draft was wrong, not the code
guiliang/cirl-benchmarks-public/utils/model_utils.py
code served (permissive licence) · get_code("e5002de1b930b693")
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 Tensor, 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

torch.manual_seed(0)

# Create a small input tensor (batch_size=2, features=4)
x = torch.randn(2, 4)

# Create an optional mask tensor with same shape
m = torch.ones(2, 4)

# Define ARGS dict with all parameters
ARGS = {
    "x": x,
    "m": m,
    "axis": -1
}
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 Tensor, 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
null_cost Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from sriram94/confidenceawareicrl
pointer only · get_code("6b3f8d5ecf6d6351")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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. No invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 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
masked_softmax 2 other papers
2 dated earlier
1908.10909 (2019-08)
Interactive Language Learning by Question Answering
2010.03768 (2020-10)
ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
null_cost 1 other paper
1 later
2406.16782 (2024-06)
Confidence Aware Inverse Constrained Reinforcement Learning

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

When deploying Reinforcement Learning (RL) agents into a physical system, we must ensure that these agents are well aware of the underlying constraints. In many real-world problems, however, the constraints are often hard to specify mathematically and unknown to the RL agents. To tackle these issues, Inverse Constrained Reinforcement Learning (ICRL) empirically estimates constraints from expert demonstrations. As an emerging research topic, ICRL does not have common benchmarks, and previous works tested algorithms under hand-crafted environments with manually-generated expert demonstrations. In this paper, we construct an ICRL benchmark in the context of RL application domains, including robot control, and autonomous driving. For each environment, we design relevant constraints and train expert agents to generate demonstration data. Besides, unlike existing baselines that learn a deterministic constraint, we propose a variational ICRL method to model a posterior distribution of candidate constraints. We conduct extensive experiments on these algorithms under our benchmark and show how they can facilitate studying important research challenges for ICRL. The benchmark, including the instructions for reproducing ICRL algorithms, is available at https://github.com/Guiliang/ICRL-benchmarks-public.

For agents

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

get_harvested_code_for_paper("2206.09670")
get_code_for_paper("2206.09670")
have("2206.09670")

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

curl https://syntology.ai/api/ran/2206.09670.json

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

[![Syntology run record](https://syntology.ai/api/ran/2206.09670.svg)](https://syntology.ai/paper/2206.09670)

Connect an agent — have() is free.