SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.09686 · 2020

Estimating means of bounded random variables by betting

arXiv · PDF · Open in the Atlas

Code that ran

8 of 9 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 6 have no contract verdict; 2 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/2010.09686.json · badge: /api/ran/2010.09686.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
wannabesmith/confseq canonical (classifier label)
classifier basis: census
8 ran · 1 no recorded run
FunctionStatusWhere it lives
banco Ran
ran on a synthesized input
wannabesmith/confseq/src/confseq/other_bounded.py
code served (permissive licence) · get_code("9478d304b69b2592")
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 tuple, ndim 2, 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

ARGS = {
    "x": [0.1, 0.2, 0.3, 0.4, 0.5],
    "alpha": 0.05,
    "running_intersection": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, 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
check_approx_equal Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
wannabesmith/confseq/demo/ate_demo.py
code served (permissive licence) · get_code("1b969d6c5e72286a")
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 bool, ndim 0, dtype family float, shape () 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 numpy as np
import pandas as pd
import torch

# Analyze the function to understand what it needs:
# - name: string
# - actual: numpy array or pandas Series with .max() method and indexing
# - expected: numpy array or pandas Series with .max() method and indexing
# - tolerance: float (default 0.0001)

# The function calls:
# - np.abs() on (expected - actual)
# - .max() on the result
# - .to_numpy() on abs_errors (so it expects a pandas Series)
# - .nonzero()[0] on a boolean array
# - indexing with error_indexes on expected and actual

# Create test data as pandas Series
torch.manual_seed(0)

# Create small pandas Series with some differences within and outside tolerance
expected_data = pd.Series([1.0, 2.0, 3.0, 4.0, 5.0])
actual_data = pd.Series([1.00005, 2.0001, 3.05, 4.0, 5.0])

ARGS = {
    'name': 'test_values',
    'actual': actual_data,
    'expected': expected_data,
    'tolerance': 0.0001
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, pandas, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () 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
form_predictions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
wannabesmith/confseq/demo/ate_demo.py
code served (permissive licence) · get_code("bb009376bd6e9abd")
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 (4,) 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 numpy as np

# Analyze the function:
# - outcomes: used in multiplication with selector, needs to be array-like
# - selector: used in multiplication with outcomes and cumsum, needs to be array-like
# - default_prediction: scalar value used to set lagged_means[0]

# Create small numpy arrays for outcomes and selector
outcomes = np.array([0.5, 0.7, 0.3, 0.9])
selector = np.array([1.0, 1.0, 0.0, 1.0])
default_prediction = 0.1

ARGS = {
    "outcomes": outcomes,
    "selector": selector,
    "default_prediction": default_prediction
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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 (4,) — shape (4,) 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
get_ci_seq Ran
ran on a synthesized input
wannabesmith/confseq/src/confseq/misc.py
code served (permissive licence) · get_code("fe264cc324a7d269")
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 tuple, ndim 2, 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 multiprocess

def ci_fn(x):
    """
    Simple confidence interval function that takes an array and returns (lower, upper) tuple.
    """
    if len(x) == 0:
        return (0.0, 1.0)
    mean = np.mean(x)
    # Simple CI based on mean
    lower = max(0.0, mean - 0.1)
    upper = min(1.0, mean + 0.1)
    return (lower, upper)

# Create setup arguments
x = np.array([0.2, 0.3, 0.5, 0.4, 0.6, 0.7, 0.8, 0.9])
ci_fn_arg = ci_fn
times = np.array([1, 2, 4, 6, 8])
parallel = False

ARGS = {
    "x": x,
    "ci_fn": ci_fn_arg,
    "times": times,
    "parallel": parallel
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
multiprocess, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, 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
lambda_aKelly Ran
ran on a synthesized input
wannabesmith/confseq/src/confseq/betting_strategies.py
code served (permissive licence) · get_code("eb94bee85d30cd36")
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 ndarray, ndim 1, dtype family float, shape (5,) 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 numpy as np

# Construct valid arguments for lambda_aKelly
x = np.array([0.5, 0.6, 0.4, 0.55, 0.65])  # sequence of observations
m = 0.5  # target mean
prior_mean = 0.5  # prior mean (default)
prior_variance = 0.25  # prior variance (default)
fake_obs = 1  # fake observations (default)
N = 100  # total number of observations (optional, set to non-None)
trunc_scale = 1.0  # truncation scale (default)

ARGS = {
    "x": x,
    "m": m,
    "prior_mean": prior_mean,
    "prior_variance": prior_variance,
    "fake_obs": fake_obs,
    "N": N,
    "trunc_scale": trunc_scale,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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 (5,) — shape (5,) 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
lambda_predmix_eb Ran
ran on a synthesized input
wannabesmith/confseq/src/confseq/betting_strategies.py
code served (permissive licence) · get_code("be384674b7b24b3a")
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 ndarray, ndim 1, dtype family float, shape (8,) 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 math
import numpy as np

# Construct valid arguments for lambda_predmix_eb
np.random.seed(0)

# Generate observed data: array-like of 0-1 bounded reals
x = np.array([0.2, 0.5, 0.3, 0.7, 0.4, 0.6, 0.1, 0.8])

ARGS = {
    "x": x,
    "truncation": math.inf,
    "alpha": 0.05,
    "fixed_n": None,
    "prior_mean": 1 / 2,
    "prior_variance": 1 / 4,
    "fake_obs": 1,
    "scale": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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 (8,) — shape (8,) 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
quantile_confidence_bound Ran
ran on a synthesized input
wannabesmith/confseq/demo/quantiles.py
code served (permissive licence) · get_code("c1c5bdedb1107baf")
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 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 math

# Analyze the function:
# - values: needs to be a sequence that can be sorted
# - p: a probability/quantile value (float between 0 and 1)
# - lower_radius: a float representing lower confidence radius
# - upper_radius: a float representing upper confidence radius

# The function sorts values and computes indices based on quantile calculations
# It returns a tuple of two confidence bounds

ARGS = {
    "values": [0.1, 0.3, 0.5, 0.7, 0.9, 0.2, 0.4, 0.6, 0.8, 0.15],
    "p": 0.5,
    "lower_radius": 0.1,
    "upper_radius": 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
rho2_opt Ran
ran on a synthesized input
wannabesmith/confseq/src/confseq/conjmix_bounded.py
code served (permissive licence) · get_code("ac05ed408f50194c")
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 float64, ndim 0, 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

# The function rho2_opt takes two parameters:
# - t_opt: int, must be >= 1
# - alpha_opt: float, must be > 0 and < 1

# Construct valid arguments
ARGS = {
    "t_opt": 5,
    "alpha_opt": 0.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
lambda_Kelly Not yet run
unverified (harvested, no recorded run)
wannabesmith/confseq/src/confseq/betting_strategies.py
code served (permissive licence) · get_code("91e9497dc3b400b7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 0 later) and 7 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.

FunctionAlso present inEarliest-datedOthers named
check_approx_equal 3 other papers
3 dated earlier
1810.08240 (2018-10)
Time-uniform, nonparametric, nonasymptotic confidence sequences
1906.09712 (2019-06)
Sequential estimation of quantiles with applications to A/B-testing and best-arm identification
2006.04347 (2020-06)
Confidence sequences for sampling without replacement
form_predictions 3 other papers
3 dated earlier
1810.08240 (2018-10)
Time-uniform, nonparametric, nonasymptotic confidence sequences
1906.09712 (2019-06)
Sequential estimation of quantiles with applications to A/B-testing and best-arm identification
2006.04347 (2020-06)
Confidence sequences for sampling without replacement

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

This paper derives confidence intervals (CI) and time-uniform confidence sequences (CS) for the classical problem of estimating an unknown mean from bounded observations. We present a general approach for deriving concentration bounds, that can be seen as a generalization and improvement of the celebrated Chernoff method. At its heart, it is based on a class of composite nonnegative martingales, with strong connections to testing by betting and the method of mixtures. We show how to extend these ideas to sampling without replacement, another heavily studied problem. In all cases, our bounds are adaptive to the unknown variance, and empirically vastly outperform existing approaches based on Hoeffding or empirical Bernstein inequalities and their recent supermartingale generalizations. In short, we establish a new state-of-the-art for four fundamental problems: CSs and CIs for bounded means, when sampling with and without replacement.

For agents

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

get_harvested_code_for_paper("2010.09686")
get_code_for_paper("2010.09686")
have("2010.09686")

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

curl https://syntology.ai/api/ran/2010.09686.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.09686.svg)](https://syntology.ai/paper/2010.09686)

Connect an agent — have() is free.