SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.08573 · NeurIPS · 2021

Neural Ensemble Search for Uncertainty Estimation and Dataset Shift

Arber Zela, Frank Hutter, Sheheryar Zaidi, Thomas Elsken, Chris Holmes, Yee Whye Teh

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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; 1 failure 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/2006.08573.json · badge: /api/ran/2006.08573.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
automl/nes canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
run_esa Ran
ran; our contract draft was wrong, not the code
automl/nes/nes/ensemble_selection/esas.py
code served (permissive licence) · get_code("81bd4ce23d0038fd")
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 torch.nn as nn

torch.manual_seed(0)

# Create a simple stand-in ESA callable
class ESA:
    def __call__(self, models_preds_pool, loss_type, M, div_strength=None):
        # Return a simple dict as output
        return {'ensemble_predictions': torch.randn(2, 3)}

# Create a stand-in model with preds attribute
class StandInModel:
    def __init__(self):
        # preds is a dict with 'val' key containing severity levels
        self.preds = {
            'val': {
                '1': torch.utils.data.TensorDataset(
                    torch.randn(2, 3),  # predictions
                    torch.randint(0, 2, (2,))  # labels
                )
            }
        }

# Build population dict with model IDs as keys
population = {
    'model_0': StandInModel(),
    'model_1': StandInModel(),
}

# Create ESA instance
esa = ESA()

# Create a simple matrix M
M = torch.randn(4, 5)

# Validation severity level
val_severity = 1

# Optional parameters
validation_size = -1
diversity_strength = None

ARGS = {
    'M': M,
    'population': population,
    'esa': esa,
    'val_severity': val_severity,
    'validation_size': validation_size,
    'diversity_strength': diversity_strength,
}
Setup block drafted by
a language model, Claude Haiku 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 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 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 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 1 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

Ensembles of neural networks achieve superior performance compared to standalone networks in terms of accuracy, uncertainty calibration and robustness to dataset shift. Deep ensembles, a state-of-the-art method for uncertainty estimation, only ensemble random initializations of a fixed architecture. Instead, we propose two methods for automatically constructing ensembles with varying architectures, which implicitly trade-off individual architectures' strengths against the ensemble's diversity and exploit architectural variation as a source of diversity. On a variety of classification tasks and modern architecture search spaces, we show that the resulting ensembles outperform deep ensembles not only in terms of accuracy but also uncertainty calibration and robustness to dataset shift. Our further analysis and ablation studies provide evidence of higher ensemble diversity due to architectural variation, resulting in ensembles that can outperform deep ensembles, even when having weaker average base learners. To foster reproducibility, our code is available: https://github.com/automl/nes * Equal contribution. 35th Conference on Neural Information Processing Systems (NeurIPS 2021).

For agents

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

get_harvested_code_for_paper("2006.08573")
get_code_for_paper("2006.08573")
have("2006.08573")

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

curl https://syntology.ai/api/ran/2006.08573.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.08573.svg)](https://syntology.ai/paper/2006.08573)

Connect an agent — have() is free.