SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.10490 · NeurIPS · 2024

Active, anytime-valid risk controlling prediction sets

Paul Mineiro, Ziyu Xu, Nikos Karampatziakis

arXiv · PDF · Open in the Atlas

Code that ran

2 of 6 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 4 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/2406.10490.json · badge: /api/ran/2406.10490.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
neilzxu/active-rcps canonical (classifier label)
classifier basis: classifier_2025
2 ran · 4 no recorded run
FunctionStatusWhere it lives
load_data Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
neilzxu/active-rcps/visualize.py
code served (permissive licence) · get_code("3d04cdca4210a66e")
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 tempfile
import os

torch.manual_seed(0)

# Create a temporary directory for the mock data
temp_dir = tempfile.mkdtemp()

# Create a minimal mock load_data module
mock_load_data_code = '''
def load_imagenet_torch_preds(data_dir):
    """Mock function that returns sample ImageNet predictions data"""
    import torch
    import pandas as pd
    
    # Return a simple dict with mock data structure
    # This would typically contain model predictions and labels
    return {
        'predictions': torch.randn(2, 1000),  # 2 samples, 1000 ImageNet classes
        'labels': torch.tensor([0, 1]),
        'metadata': pd.DataFrame({
            'image_id': ['img_0', 'img_1'],
            'class': [0, 1]
        })
    }
'''

# Write the mock module to a temporary location
mock_module_path = os.path.join(temp_dir, 'load_data.py')
with open(mock_module_path, 'w') as f:
    f.write(mock_load_data_code)

# Add temp directory to sys.path so the mock module can be imported
import sys
sys.path.insert(0, temp_dir)

ARGS = {
    'data_dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
load_data, matplotlib, numpy, pandas, seaborn, torch, tqdm — 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
predicted_risk Ran
ran, honoured the contract we drafted
neilzxu/active-rcps/visualize.py
code served (permissive licence) · get_code("6ade24d92ea34b4b")
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 ndarray, ndim 1, dtype family float, shape (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 numpy as np

torch.manual_seed(0)

# P: tensor of shape (batch_size, num_features)
# The function uses P with indexing P[i, :] and operations like P * w.reshape(1, -1)
P = torch.randn(2, 4)

# w: tensor of shape (num_features,)
# The function uses w.reshape(1, -1), w.shape[0], and indexing w[exc_idxs]
w = torch.randn(4)

# beta: scalar threshold value
# The function compares cumulative sums against beta
beta = 2.5

ARGS = {
    "P": P,
    "w": w,
    "beta": beta,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
matplotlib, numpy, pandas, seaborn, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (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
BaseMartingale Not yet run
unverified (harvested, no recorded run)
neilzxu/active-rcps/IwUpperMartingale.py
code served (permissive licence) · get_code("b59d668388ec8cb9")
How it ran

no invocation recorded.

FullIwUpperMartingale Not yet run
unverified (harvested, no recorded run)
neilzxu/active-rcps/IwUpperMartingale.py
code served (permissive licence) · get_code("f38ed9c4493769eb")
How it ran

no invocation recorded.

GridMartingale Not yet run
unverified (harvested, no recorded run)
neilzxu/active-rcps/IwUpperMartingale.py
code served (permissive licence) · get_code("c50dffc7f637d9a5")
How it ran

no invocation recorded.

plot_imagenet_true_rho Not yet run
unverified (harvested, no recorded run)
neilzxu/active-rcps/visualize.py
code served (permissive licence) · get_code("76d6ae47bdfeffa6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 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 6 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

Rigorously establishing the safety of black-box machine learning models concerning critical risk measures is important for providing guarantees about model behavior. Recently, Bates et. al. (JACM '24) introduced the notion of a risk controlling prediction set (RCPS) for producing prediction sets that are statistically guaranteed low risk from machine learning models. Our method extends this notion to the sequential setting, where we provide guarantees even when the data is collected adaptively, and ensures that the risk guarantee is anytime-valid, i.e., simultaneously holds at all time steps. Further, we propose a framework for constructing RCPSes for active labeling, i.e., allowing one to use a labeling policy that chooses whether to query the true label for each received data point and ensures that the expected proportion of data points whose labels are queried are below a predetermined label budget. We also describe how to use predictors (i.e., the machine learning model for which we provide risk control guarantees) to further improve the utility of our RCPSes by estimating the expected risk conditioned on the covariates. We characterize the optimal choices of label policy and predictor under a fixed label budget and show a regret result that relates the estimation error of the optimal labeling policy and predictor to the wealth process that underlies our RCPSes. Lastly, we present practical ways of formulating label policies and empirically show that our label policies use fewer labels to reach higher utility than naive baseline labeling strategies on both simulations and real data.

For agents

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

get_harvested_code_for_paper("2406.10490")
get_code_for_paper("2406.10490")
have("2406.10490")

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

curl https://syntology.ai/api/ran/2406.10490.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.10490.svg)](https://syntology.ai/paper/2406.10490)

Connect an agent — have() is free.