SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.11670 · 2022

Wide Mean-Field Bayesian Neural Networks Ignore the Data

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 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 0 have no contract verdict; 3 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/2202.11670.json · badge: /api/ran/2202.11670.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
dtak/wide-bnns-public canonical (classifier label)
classifier basis: classifier_2025
3 ran · 1 no recorded run
FunctionStatusWhere it lives
plot_predictive Ran
ran; our contract draft was wrong, not the code
dtak/wide-bnns-public/experiments/figures/work/make_figures.py
code served (permissive licence) · get_code("a0a0864fe2c66299")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned tuple, 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 numpy as np
import matplotlib.pyplot as plt

np.random.seed(0)

# Create a grid of x values
n_grid = 50
x_grid = np.linspace(0, 10, n_grid).reshape(-1, 1)

# Create samples from a predictive distribution
n_samp = 20
f_grid = np.random.randn(n_samp, n_grid) * 0.5 + np.sin(x_grid.T)

# Create a true function
f_true = np.sin(x_grid)

# Create a figure and axis
fig, ax = plt.subplots()

ARGS = {
    'x_grid': x_grid,
    'f_grid': f_grid,
    'uncertainty_type': 'std',
    'n_std': 2,
    'n_samp_show': 3,
    'f_true': f_true,
    'color': 'blue',
    'label': 'prediction',
    'ax': ax
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
matplotlib, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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
plot_upcrossings Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
dtak/wide-bnns-public/experiments/figures/work/make_figures.py
code served (permissive licence) · get_code("6a2fcddfac274eb4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Axes. 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
import matplotlib.pyplot as plt

np.random.seed(0)

# Create sorted x values
n = 20
x = np.linspace(0, 10, n)

# Create f with multiple samples
n_samp = 50
f = np.sin(x) + np.random.randn(n_samp, n) * 0.3

# Set upcross_level
upcross_level = 0.0

# Set bins
bins = 8

# Create an axes object
fig, ax = plt.subplots()

ARGS = {
    'x': x,
    'f': f,
    'upcross_level': upcross_level,
    'bins': bins,
    'ax': ax
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
matplotlib, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Axes — a Axes output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Axes 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
scalable_foward Ran
ran; our contract draft was wrong, not the code
dtak/wide-bnns-public/scripts/experiment_pytorch.py
code served (permissive licence) · get_code("e9c63691c59a1449")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Tensor, ndim 3, dtype family float, shape (3, 1000, 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 torch
import torch.nn as nn

torch.manual_seed(0)

class MockBNN(nn.Module):
    def __init__(self, dim_hidden):
        super().__init__()
        self.dim_hidden = dim_hidden
        self.linear = nn.Linear(5, 3)
    
    def forward_batched(self, x, n_samp=1000, n_batch=100, prior=False):
        # x shape: (batch_size, input_dim)
        # Returns shape: (batch_size, n_samp, output_dim) or similar
        batch_size = x.shape[0]
        output_dim = 3
        # Return tensor of shape (batch_size, n_samp, output_dim)
        return torch.randn(batch_size, n_samp, output_dim)

# Create a BNN with dim_hidden below threshold
bnn = MockBNN(dim_hidden=100000)

# Create input tensor with shape (2, 1, 5) to match the indexing x[i, 0]
x = torch.randn(2, 1, 5)

ARGS = {
    'bnn': bnn,
    'x': x,
    'n_samp': 1000,
    'prior': False,
    'dim_hidden_thresh': 256000
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 3, dtype family float, shape (3, 1000, 2) — shape (3, 1000, 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
compute_metrics Not yet run
unverified (harvested, no recorded run)
dtak/wide-bnns-public/scripts/experiment_pytorch.py
code served (permissive licence) · get_code("697d6e902861b9c5")
How it ran

no invocation recorded.

Also present in other papers' code

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

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

Bayesian neural networks (BNNs) combine the expressive power of deep learning with the advantages of Bayesian formalism. In recent years, the analysis of wide, deep BNNs has provided theoretical insight into their priors and posteriors. However, we have no analogous insight into their posteriors under approximate inference. In this work, we show that mean-field variational inference entirely fails to model the data when the network width is large and the activation function is odd. Specifically, for fully-connected BNNs with odd activation functions and a homoscedastic Gaussian likelihood, we show that the optimal mean-field variational posterior predictive (i.e., function space) distribution converges to the prior predictive distribution as the width tends to infinity. We generalize aspects of this result to other likelihoods. Our theoretical results are suggestive of underfitting behavior previously observered in BNNs. While our convergence bounds are non-asymptotic and constants in our analysis can be computed, they are currently too loose to be applicable in standard training regimes. Finally, we show that the optimal approximate posterior need not tend to the prior if the activation function is not odd, showing that our statements cannot be generalized arbitrarily.

For agents

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

get_harvested_code_for_paper("2202.11670")
get_code_for_paper("2202.11670")
have("2202.11670")

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

curl https://syntology.ai/api/ran/2202.11670.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.11670.svg)](https://syntology.ai/paper/2202.11670)

Connect an agent — have() is free.