SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.13536 · NeurIPS · 2022

LobsDICE: Offline Learning from Observation via Stationary Distribution Correction Estimation

Hongseok Yang, Geon-Hyeong Kim, Kee-Eung Kim, Jongmin Lee, Youngsoo Jang

arXiv · PDF · Open in the Atlas

Code that ran

2 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; 0 failures of our instrument; 3 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/2202.13536.json · badge: /api/ran/2202.13536.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
geon-hyeong/imitation-dice canonical (classifier label)
classifier basis not recorded on the link
2 ran · 3 no recorded run
FunctionStatusWhere it lives
add_absorbing_states Ran
ran on a synthesized input
geon-hyeong/imitation-dice/utils.py
code served (permissive licence) · get_code("de7dfe13ea6c9e3d")
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

class MockActionSpaceDiscrete:
    def __init__(self):
        self.n = 3
        self.dtype = int

class MockActionSpaceContinuous:
    def __init__(self):
        self.shape = (2,)
        self.dtype = float

class MockEnv:
    def __init__(self):
        self.action_space = MockActionSpaceDiscrete()
        self._max_episode_steps = 10
    
    def get_absorbing_state(self):
        return np.array([0.0, 0.0, 0.0, 0.0, 1.0])

expert_states = np.random.randn(4, 4).astype(np.float32)
expert_actions = np.random.randn(4, 3).astype(np.float32)
expert_next_states = np.random.randn(4, 4).astype(np.float32)
expert_dones = np.array([0.0, 1.0, 0.0, 1.0], dtype=np.float32)
env = MockEnv()

ARGS = {
    "expert_states": expert_states,
    "expert_actions": expert_actions,
    "expert_next_states": expert_next_states,
    "expert_dones": expert_dones,
    "env": env,
    "dtype": np.float32
}
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
check_and_normalize_box_actions Ran
ran on a synthesized input
geon-hyeong/imitation-dice/wrappers/normalize_action_wrapper.py
code served (permissive licence) · get_code("b1c0ce2e87ab7757")
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 NormalizeBoxActionWrapper. 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 gym
from gym import spaces
import numpy as np

# Create a simple environment with Box action space
# that has action bounds different from [-1, 1]
class SimpleEnv(gym.Env):
    def __init__(self):
        self.action_space = spaces.Box(low=0.0, high=10.0, shape=(3,), dtype=np.float32)
        self.observation_space = spaces.Box(low=-np.inf, high=np.inf, shape=(4,), dtype=np.float32)
        self._max_episode_steps = 100
    
    def reset(self):
        return self.observation_space.sample()
    
    def step(self, action):
        observation = self.observation_space.sample()
        reward = 0.0
        done = False
        info = {}
        return observation, reward, done, info

# Create an environment instance
env = SimpleEnv()

ARGS = {
    'env': env
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
absl, gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type NormalizeBoxActionWrapper — a NormalizeBoxActionWrapper output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NormalizeBoxActionWrapper 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
boolean Not yet run
unverified (harvested, no recorded run)
geon-hyeong/imitation-dice/config/lfd_default_config.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ZhengYinan-AIR/OMIGA, is pointer only (licence: NONE) · get_code("3b81875ee24223a6")
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 ZhengYinan-AIR/OMIGA, 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. No invocation recorded.

evaluate_d4rl Not yet run
unverified (harvested, no recorded run)
geon-hyeong/imitation-dice/lfd_mujoco.py
code served (permissive licence) · get_code("e2cd2e34acacdb21")
How it ran

no invocation recorded.

load_d4rl_data Not yet run
unverified (harvested, no recorded run)
geon-hyeong/imitation-dice/utils.py
code served (permissive licence) · get_code("8a5275eeca2fbda2")
How it ran

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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 4 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
boolean 5 other papers
5 later
2302.01667 (2023-02)
Mind the Gap: Offline Policy Optimization for Imperfect Rewards
2307.11620 (2023-07)
Offline Multi-Agent Reinforcement Learning with Implicit Global-to-Local Value Regularization
2405.17477 (2024-05)
OLLIE: Imitation Learning from Offline Pretraining to Online Finetuning
and 2 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 2026-11-22.

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 consider the problem of learning from observation (LfO), in which the agent aims to mimic the expert's behavior from the state-only demonstrations by experts. We additionally assume that the agent cannot interact with the environment but has access to the action-labeled transition data collected by some agents with unknown qualities. This offline setting for LfO is appealing in many real-world scenarios where the ground-truth expert actions are inaccessible and the arbitrary environment interactions are costly or risky. In this paper, we present LobsDICE, an offline LfO algorithm that learns to imitate the expert policy via optimization in the space of stationary distributions. Our algorithm solves a single convex minimization problem, which minimizes the divergence between the two statetransition distributions induced by the expert and the agent policy. Through an extensive set of offline LfO tasks, we show that LobsDICE outperforms strong baseline methods. * Equal contribution. † Work done while the authors were students at KAIST. 36th Conference on Neural Information Processing Systems (NeurIPS 2022).

For agents

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

get_harvested_code_for_paper("2202.13536")
get_code_for_paper("2202.13536")
have("2202.13536")

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

curl https://syntology.ai/api/ran/2202.13536.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.13536.svg)](https://syntology.ai/paper/2202.13536)

Connect an agent — have() is free.