SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.15871 · NeurIPS · 2023

Learning Robust Statistics for Simulation-based Inference under Model Misspecification

Luigi Acerbi, Amauri Souza, Samuel Kaski, Ayush Bharti, Daolang Huang

arXiv · PDF · Open in the Atlas

Code that ran

12 of 12 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; 10 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/2305.15871.json · badge: /api/ran/2305.15871.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
mackelab/sbi canonical (classifier label)
classifier basis: census
8 ran · 0 no recorded run
huangdaolang/robust-sbi canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
danielward27/rnpe canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
TurinModel Ran
ran on a synthesized input
huangdaolang/robust-sbi/simulators/turin.py
code served (permissive licence) · get_code("3d6cf48c8adb6af6")
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 (100, 801) 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

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

# Define arguments for TurinModel
ARGS = {
    'G0': 1.0,                          # Path loss at reference distance
    'T': 1e-6,                          # RMS delay spread (seconds)
    'lambda_0': 10.0,                   # Average number of clusters
    'sigma2_N': 0.01,                   # Noise power
    'B': 4000000000.0,                  # Bandwidth (Hz)
    'Ns': 801,                          # Number of frequency samples
    'N': 100,                           # Number of receive antennas
    'tau0': 0                           # Delay offset (seconds)
}
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 (100, 801) — shape (100, 801) 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
add_spike_and_slab_error Ran
ran; our contract draft was wrong, not the code
danielward27/rnpe/scripts/run_task.py
code served (permissive licence) · get_code("6f99b93048e3da8d")
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 ArrayImpl, 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 jax
import jax.numpy as jnp
from jax import random

# Set seed for reproducibility
key = random.PRNGKey(0)

# Create a small input array (batch_size=2, features=4)
x = jnp.array([[1.0, 2.0, 3.0, 4.0],
               [5.0, 6.0, 7.0, 8.0]])

# Define parameters
slab_scale = 2.5
spike_scale = 0.01

ARGS = {
    "key": key,
    "x": x,
    "slab_scale": slab_scale,
    "spike_scale": spike_scale,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
compute_rbf_mmd Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/diagnostics/misspecification.py
code served (permissive licence) · get_code("fc0e53b9b168ce47")
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 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)

# Create small tensors for x and y
# x: shape (2, 4) - 2 samples, 4 dimensions
# y: shape (2, 4) - 2 samples, 4 dimensions
x = torch.randn(2, 4)
y = torch.randn(2, 4)

# bandwidth parameter
bandwidth = 1.0

# mode parameter
mode = 'biased'

ARGS = {
    'x': x,
    'y': y,
    'bandwidth': bandwidth,
    'mode': mode
}
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 Tensor, 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
eval_lc2st Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/diagnostics/lc2st.py
code served (permissive licence) · get_code("042c1c4d2e2f3e1c")
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
from sklearn.linear_model import LogisticRegression
from sklearn.base import BaseEstimator

torch.manual_seed(0)

# Create sample tensors
sample_size = 2
dim_theta = 4
dim_x = 3

theta_p = torch.randn(sample_size, dim_theta)
x_o = torch.randn(1, dim_x)

# Train a simple classifier
# The function expects a classifier with predict_proba method
# It concatenates theta_p (sample_size, dim_theta) with x_o repeated (sample_size, dim_x)
# So the joint input will be (sample_size, dim_theta + dim_x)
X_train = np.random.randn(10, dim_theta + dim_x)
y_train = np.random.randint(0, 2, 10)

clf = LogisticRegression(random_state=0, max_iter=1000)
clf.fit(X_train, y_train)

ARGS = {
    "theta_p": theta_p,
    "x_o": x_o,
    "clf": clf,
    "return_proba": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, 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
get_1d_marginal_peaks_from_kde Ran
ran on a synthesized input
huangdaolang/robust-sbi/utils/analysis_utils.py
code served (permissive licence) · get_code("3cda7949f2409864")
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 1, dtype family float, shape (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 torch
import numpy as np
from scipy.stats import gaussian_kde

torch.manual_seed(0)

# Create small sample tensor: (num_samples, num_dimensions)
# Shape (5, 3) = 5 samples, 3 dimensions
samples = torch.randn(5, 3)

ARGS = {
    'samples': samples,
    'num_candidates': 100,
    'num_workers': 1,
    'bw_method': 'scott'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
joblib, numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (3,) — shape (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
get_default_diag_kwargs Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/analysis/plotting_classes.py
code served (permissive licence) · get_code("db2fea809ee31e99")
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 matplotlib.pyplot as plt
from dataclasses import asdict, dataclass, field
from typing import Any, Dict, List, Optional

@dataclass(frozen=True)
class DiagOptions:
    """
    Base class for keyword arguments used in diagonal plots.

    This class serves as a common parent for specific diagonal plot
    configuration classes such as KDE, Histogram, and Scatter.
    """
    mpl_kwargs: Dict[str, Any] = field(default_factory=dict)

@dataclass(frozen=True)
class KdeDiagOptions(DiagOptions):
    bw_method: str = 'scott'
    bins: int = 50

@dataclass(frozen=True)
class HistDiagOptions(DiagOptions):
    bin_heuristic: str = 'Freedman-Diaconis'

    def __post_init__(self):
        mpl_kwargs_defaults = {'density': False, 'histtype': 'step'}
        updated = {**mpl_kwargs_defaults, **self.mpl_kwargs}
        object.__setattr__(self, 'mpl_kwargs', updated)

@dataclass(frozen=True)
class ScatterDiagOptions(DiagOptions):
    ...

@dataclass(frozen=True)
class BarDiagOptions(DiagOptions):
    """Options for discrete bar charts on diagonal."""
    width: float = 0.8

    def __post_init__(self):
        mpl_kwargs_defaults = {'alpha': 0.7, 'edgecolor': 'white'}
        updated = {**mpl_kwargs_defaults, **self.mpl_kwargs}
        object.__setattr__(self, 'mpl_kwargs', updated)

def _set_color(i: int) -> str:
    """
    Returns a distinct color from Matplotlib's default color cycle.

    Args:
        i: Index used to select a color, spaced by 2.

    Returns:
        str: A color string from the Matplotlib color cycle.
    """
    colors = plt.rcParams['axes.prop_cycle'].by_key()['color']
    new_color = colors[i * 2 % len(colors)]
    return new_color

def get_default_diag_kwargs(diag: Optional[str], i: int=0) -> Dict[str, Any]:
    """Get default diag kwargs."""
    if diag == 'kde':
        diag_options = KdeDiagOptions(mpl_kwargs=dict(color=_set_color(i)))
    elif diag == 'hist':
        diag_options = HistDiagOptions(mpl_kwargs=dict(color=_set_color(i)))
    elif diag == 'scatter':
        diag_options = ScatterDiagOptions(mpl_kwargs=dict(color=_set_color(i)))
    elif diag == 'bar':
        diag_options = BarDiagOptions(mpl_kwargs=dict(color=_set_color(i)))
    else:
        return {}
    return asdict(diag_options)

ARGS = {
    'diag': 'kde',
    'i': 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib — 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
get_default_offdiag_kwargs Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/analysis/plotting_classes.py
code served (permissive licence) · get_code("9e075659a09f2c50")
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 matplotlib import pyplot as plt

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

# The function get_default_offdiag_kwargs takes two parameters:
# 1. offdiag: Optional[str] - can be 'kde', 'kde2d', 'hist', 'hist2d', 'scatter', 'contour', 'contourf', 'plot', or None
# 2. i: int - defaults to 0, used as index for color selection

# Test with 'kde' offdiag type
ARGS = {
    'offdiag': 'kde',
    'i': 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, 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
get_kde Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
mackelab/sbi/sbi/analysis/plot.py
code served (permissive licence) · get_code("f75a8e2562862bbf")
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, ndim 3, 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
import torch

torch.manual_seed(0)

# Create sample data for 2D KDE
np.random.seed(0)
samples_col = np.random.randn(100)
samples_row = np.random.randn(100)

# Create limit tensors
limits_col = torch.tensor([-3.0, 3.0])
limits_row = torch.tensor([-3.0, 3.0])

# Create offdiag_kwargs dictionary with required parameters
offdiag_kwargs = {
    'bw_method': 0.2,
    'bins': 16,
    'percentile': True,
    'levels': [0.1, 0.5, 0.9]
}

ARGS = {
    'samples_col': samples_col,
    'samples_row': samples_row,
    'limits_col': limits_col,
    'limits_row': limits_row,
    'offdiag_kwargs': offdiag_kwargs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, dtype family float — a tuple 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
median_heuristic Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/diagnostics/misspecification.py
code served (permissive licence) · get_code("982bf515ea600367")
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
from torch import Tensor

torch.manual_seed(0)

# Create two small tensors with distinct dimensions
# x: batch size 2, dimension 4
# y: batch size 3, dimension 4
x = torch.randn(2, 4)
y = torch.randn(3, 4)

ARGS = {
    "x": x,
    "y": y,
}
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
permute_data Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/diagnostics/lc2st.py
code served (permissive licence) · get_code("4b99ed3c2e6b6613")
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, ndim 3, 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
from torch import Tensor

torch.manual_seed(0)

# Create small sample tensors
# theta_p: samples from P, shape (sample_size, dim)
# theta_q: samples from Q, shape (sample_size, dim)
sample_size = 2
dim = 4

theta_p = torch.randn(sample_size, dim)
theta_q = torch.randn(sample_size, dim)
seed = 1

ARGS = {
    "theta_p": theta_p,
    "theta_q": theta_q,
    "seed": seed,
}
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, ndim 3, dtype family float — a tuple 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
rbf_kernel Ran
ran; our contract draft was wrong, not the code
mackelab/sbi/sbi/diagnostics/misspecification.py
code served (permissive licence) · get_code("62ed8bd9e7daa553")
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, 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 torch

torch.manual_seed(0)

# Create small tensors for x and y
# x: (2, 4) - 2 samples, 4 dimensions
# y: (3, 4) - 3 samples, 4 dimensions (distinct size to disambiguate)
x = torch.randn(2, 4)
y = torch.randn(3, 4)
bandwidth = 1.0

ARGS = {
    "x": x,
    "y": y,
    "bandwidth": bandwidth,
}
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, 3) — shape (2, 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
rescale_results Ran
ran; our contract draft was wrong, not the code
danielward27/rnpe/scripts/run_task.py
code served (permissive licence) · get_code("90853ad231944023")
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.

import torch
import numpy as np

torch.manual_seed(0)

# Construct the res dictionary with all required nested structure
res = {
    'scales': {
        'x_mean': torch.tensor([1.0, 2.0, 3.0, 4.0]),
        'x_std': torch.tensor([0.5, 0.6, 0.7, 0.8]),
        'theta_mean': torch.tensor([10.0, 20.0, 30.0]),
        'theta_std': torch.tensor([2.0, 3.0, 4.0])
    },
    'data': {
        'x': torch.randn(2, 4),
        'y': torch.randn(2, 4),
        'theta': torch.randn(2, 3),
        'theta_true': torch.randn(2, 3)
    },
    'mcmc_samples': {
        'x': torch.randn(2, 4)
    },
    'posterior_samples': {
        'NPE': torch.randn(2, 3),
        'RNPE': torch.randn(2, 3)
    }
}

ARGS = {
    'res': res
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 8 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 8 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
permute_data 3 other papers
3 later
2404.07593 (2024-04)
Diffusion posterior sampling for simulation-based inference in tall data settings
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
2607.05252 (2026-07)
FUSE: FK-Steered Multi-Modal Flow Matching for Efficient Simulation-Based Posterior Estimation
compute_rbf_mmd 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
eval_lc2st 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
get_default_diag_kwargs 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
get_default_offdiag_kwargs 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
get_kde 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
median_heuristic 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows
rbf_kernel 1 other paper
1 later
2411.17337 (2024-11)
sbi reloaded: a toolkit for simulation-based inference workflows

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

Simulation-based inference (SBI) methods such as approximate Bayesian computation (ABC), synthetic likelihood, and neural posterior estimation (NPE) rely on simulating statistics to infer parameters of intractable likelihood models. However, such methods are known to yield untrustworthy and misleading inference outcomes under model misspecification, thus hindering their widespread applicability. In this work, we propose the first general approach to handle model misspecification that works across different classes of SBI methods. Leveraging the fact that the choice of statistics determines the degree of misspecification in SBI, we introduce a regularized loss function that penalises those statistics that increase the mismatch between the data and the model. Taking NPE and ABC as use cases, we demonstrate the superior performance of our method on high-dimensional time-series models that are artificially misspecified. We also apply our method to real data from the field of radio propagation where the model is known to be misspecified. We show empirically that the method yields robust inference in misspecified scenarios, whilst still being accurate when the model is well-specified.

For agents

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

get_harvested_code_for_paper("2305.15871")
get_code_for_paper("2305.15871")
have("2305.15871")

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

curl https://syntology.ai/api/ran/2305.15871.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.15871.svg)](https://syntology.ai/paper/2305.15871)

Connect an agent — have() is free.