SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.03260 · 2019

Reliable training and estimation of variance networks

arXiv · PDF · Open in the Atlas

Code that ran

8 of 10 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; 8 failures of our instrument; 2 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/1906.03260.json · badge: /api/ran/1906.03260.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
SkafteNicki/john canonical (classifier label)
classifier basis: classifier_2025
3 ran · 1 no recorded run
astirn/variational-variance extension (classifier label)
classifier basis: classifier_2025
5 ran · 1 no recorded run
FunctionStatusWhere it lives
drop_detlefsen Ran
ran; our contract draft was wrong, not the code
astirn/variational-variance/regression_analysis.py
code served (permissive licence) · get_code("88e013a94fe15c9f")
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 DataFrame. 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 pandas as pd

df = pd.DataFrame({
    'Algorithm': ['Detlefsen', 'Other', 'Detlefsen', 'Another'],
    'Value': [1, 2, 3, 4]
})

ARGS = {
    'df': df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
gen_Qw Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SkafteNicki/john/locality_sampler.py
pointer only (licence: NONE) · get_code("1212679baefb9b71")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, SkafteNicki/john 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. A run is recorded.

integrate_active_learning_curves Ran
ran; our contract draft was wrong, not the code
astirn/variational-variance/active_learning_analysis.py
code served (permissive licence) · get_code("93b3ea16c3cdd175")
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 DataFrame. 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 pandas as pd
import numpy as np

# Create a sample DataFrame that mimics an active learning log
# The function groups by 'Algorithm', 'Prior', and the index, then sums a metric
data = {
    'Algorithm': ['algo1', 'algo1', 'algo2', 'algo2'],
    'Prior': ['prior_a', 'prior_a', 'prior_b', 'prior_b'],
    'accuracy': [0.85, 0.90, 0.78, 0.88],
    'loss': [0.15, 0.10, 0.22, 0.12]
}

log = pd.DataFrame(data)

# Set a meaningful index for groupby operation
log.index = pd.Index([0, 1, 0, 1], name='iteration')

ARGS = {
    'log': log,
    'metric': 'accuracy'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
keep_fashion Ran
ran; our contract draft was wrong, not the code
astirn/variational-variance/generative_analysis.py
code served (permissive licence) · get_code("6170c011a2239b60")
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 DataFrame. 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 pandas as pd

torch.manual_seed(0)

# The function filters a pandas DataFrame by the 'Dataset' column
# looking for rows where Dataset == 'fashion mnist'
# So we need a DataFrame with a 'Dataset' column

results = pd.DataFrame({
    'Dataset': ['fashion mnist', 'cifar10', 'fashion mnist', 'imagenet'],
    'accuracy': [0.95, 0.87, 0.92, 0.98],
    'loss': [0.1, 0.3, 0.15, 0.05]
})

ARGS = {
    'results': results
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
keep_mnist Ran
ran; our contract draft was wrong, not the code
astirn/variational-variance/generative_analysis.py
code served (permissive licence) · get_code("f6fa0ca68d2fe1e2")
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 DataFrame. 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 pandas as pd

# The function filters a DataFrame by the 'Dataset' column looking for 'mnist'
# We need to create a DataFrame with a 'Dataset' column

results = pd.DataFrame({
    'Dataset': ['mnist', 'cifar10', 'mnist', 'imagenet'],
    'Accuracy': [0.95, 0.87, 0.92, 0.78],
    'Loss': [0.12, 0.34, 0.15, 0.45]
})

ARGS = {
    'results': results
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
locality_sampler Ran
ran; our contract draft was wrong, not the code
SkafteNicki/john/locality_sampler.py
pointer only (licence: NONE) · get_code("142eaf6178104a3f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, SkafteNicki/john 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. A run is recorded.

locality_sampler2 Ran
ran; our contract draft was wrong, not the code
SkafteNicki/john/locality_sampler.py
pointer only (licence: NONE) · get_code("6cafd798a1a01685")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, SkafteNicki/john 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. A run is recorded.

string_table Ran
ran; our contract draft was wrong, not the code
astirn/variational-variance/generative_analysis.py
code served (permissive licence) · get_code("2a8e9acf7f3436c8")
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 DataFrame. 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 pandas as pd
import numpy as np

# Create a small DataFrame with numeric values
# Some values > 0.1 (formatted as .2f) and some < 0.1 (formatted as .1e)
df = pd.DataFrame({
    'col1': [0.05, 0.25, -0.03, 1.5],
    'col2': [0.001, 0.99, -0.0001, 2.3],
    'col3': [0.15, -0.08, 0.5, 0.002]
})

ARGS = {
    'df': df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
gp Not yet run
unverified (harvested, no recorded run)
SkafteNicki/john/experiment_active_learning.py
pointer only (licence: NONE) · get_code("c858f668b539babc")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, SkafteNicki/john 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.

uci_reported Not yet run
unverified (harvested, no recorded run)
astirn/variational-variance/regression_analysis.py
code served (permissive licence) · get_code("818d4dd7e1be914f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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

We propose and investigate new complementary methodologies for estimating predictive variance networks in regression neural networks. We derive a locally aware mini-batching scheme that result in sparse robust gradients, and show how to make unbiased weight updates to a variance network. Further, we formulate a heuristic for robustly fitting both the mean and variance networks post hoc. Finally, we take inspiration from posterior Gaussian processes and propose a network architecture with similar extrapolation properties to Gaussian processes. The proposed methodologies are complementary, and improve upon baseline methods individually. Experimentally, we investigate the impact on predictive uncertainty on multiple datasets and tasks ranging from regression, active learning and generative modeling. Experiments consistently show significant improvements in predictive uncertainty estimation over state-of-the-art methods across tasks and datasets.

For agents

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

get_harvested_code_for_paper("1906.03260")
get_code_for_paper("1906.03260")
have("1906.03260")

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

curl https://syntology.ai/api/ran/1906.03260.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.03260.svg)](https://syntology.ai/paper/1906.03260)

Connect an agent — have() is free.