SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.13570 · NeurIPS · 2020

Hyperparameter Ensembles for Robustness and Uncertainty Quantification

Jasper Snoek, Rodolphe Jenatton, Florian Wenzel, Dustin Tran

arXiv · PDF · Open in the Atlas

Code that ran

2 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 1 failure 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/2006.13570.json · badge: /api/ran/2006.13570.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
google/uncertainty-baselines extension (classifier label)
classifier basis: classifier_2025
2 ran · 2 no recorded run
FunctionStatusWhere it lives
LogScaler Ran
ran on a synthesized input · constructed an object; no result was computed
google/uncertainty-baselines/uncertainty_baselines/models/wide_resnet_hyperbatchensemble.py
code served (permissive licence) · get_code("109f2e00c01a3c25")
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 an instance of its own class, LogScaler: the run constructed the object and did not call it, so no result was computed. 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 tensorflow as tf

# Define ranges parameter - a list of dicts with 'min' and 'max' keys
ranges = [
    {'min': 0.1, 'max': 1.0},
    {'min': 0.01, 'max': 0.5},
    {'min': 1.0, 'max': 10.0},
    {'min': 0.5, 'max': 5.0}
]

ARGS = {
    'ranges': ranges
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type LogScaler — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a LogScaler 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_sweep Ran
ran; our contract draft was wrong, not the code
google/uncertainty-baselines/baselines/cifar/hyperdeepensemble_configs/cifar10_rand_search_sweep.py
code served (permissive licence) · get_code("8c4984235cb0593e")
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

class MockHyper:
    """Mock hyperparameter search object that mimics the hyper API."""
    
    def __init__(self):
        self.counter = 0
    
    def fixed(self, name, value, length=1):
        """Return a fixed hyperparameter."""
        return {'type': 'fixed', 'name': name, 'value': value, 'length': length}
    
    def loguniform(self, name, interval):
        """Return a loguniform hyperparameter."""
        return {'type': 'loguniform', 'name': name, 'interval': interval, 'length': 1}
    
    def uniform(self, name, domain):
        """Return a uniform hyperparameter."""
        return {'type': 'uniform', 'name': name, 'domain': domain, 'length': 1}
    
    def interval(self, low, high):
        """Return an interval."""
        return {'low': low, 'high': high}
    
    def discrete(self, values):
        """Return a discrete domain."""
        return list(values)
    
    def zipit(self, hps, length=1):
        """Zip hyperparameters together."""
        return {'type': 'zipit', 'hps': hps, 'length': length}
    
    def chainit(self, iterables):
        """Chain iterables together."""
        return {'type': 'chainit', 'iterables': iterables}

ARGS = {
    'hyper': MockHyper()
}
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
make_sign_initializer Not yet run
unverified (harvested, no recorded run)
google/uncertainty-baselines/uncertainty_baselines/models/wide_resnet_hyperbatchensemble.py
code served (permissive licence) · get_code("1c87402304aeab38")
How it ran

no invocation recorded.

wide_resnet_hyperbatchensemble Not yet run
unverified (harvested, no recorded run)
google/uncertainty-baselines/uncertainty_baselines/models/wide_resnet_hyperbatchensemble.py
code served (permissive licence) · get_code("99ee23e47179ff60")
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

Ensembles over neural network weights trained from different random initialization, known as deep ensembles, achieve state-of-the-art accuracy and calibration. The recently introduced batch ensembles provide a drop-in replacement that is more parameter efficient. In this paper, we design ensembles not only over weights, but over hyperparameters to improve the state of the art in both settings. For best performance independent of budget, we propose hyper-deep ensembles, a simple procedure that involves a random search over different hyperparameters, themselves stratified across multiple random initializations. Its strong performance highlights the benefit of combining models with both weight and hyperparameter diversity. We further propose a parameter efficient version, hyper-batch ensembles, which builds on the layer structure of batch ensembles and self-tuning networks. The computational and memory costs of our method are notably lower than typical ensembles. On image classification tasks, with MLP, LeNet, ResNet 20 and Wide ResNet 28-10 architectures, we improve upon both deep and batch ensembles.

For agents

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

get_harvested_code_for_paper("2006.13570")
get_code_for_paper("2006.13570")
have("2006.13570")

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

curl https://syntology.ai/api/ran/2006.13570.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.13570.svg)](https://syntology.ai/paper/2006.13570)

Connect an agent — have() is free.