SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.01948 · 2022

Game-theoretic statistics and safe anytime-valid inference

arXiv · PDF · Open in the Atlas

Code that ran

12 of 15 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 12 have no contract verdict; 0 failures of our instrument; 3 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/2210.01948.json · badge: /api/ran/2210.01948.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
yjchoe/ComparingForecasters community (classifier label)
classifier basis not recorded on the link
12 ran · 3 no recorded run
FunctionStatusWhere it lives
calibrate_p_to_e Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/eprocess.py
code served (permissive licence) · get_code("537f3e7375564e66")
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 (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

# Create test data with values in [0, 1]
ps = np.array([0.1, 0.5, 0.9, 0.3])

ARGS = {
    'ps': ps,
    'strategy': 'mixture'
}
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 (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
cgf_bernoulli Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/cgfs.py
code served (permissive licence) · get_code("76b12d1a2a48d433")
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; 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 = {
    'lambdas': np.array([0.5, 1.0, 1.5, 2.0]),
    'g': 0.8,
    'h': 1.2,
}
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 — no output shape is stated: observe_signature.py could not observe it (no_result)
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
cgf_gaussian Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/cgfs.py
code served (permissive licence) · get_code("4f93ae36d9f0d15f")
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 (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

# The function cgf_gaussian takes lambdas as ArrayLike and **kwargs
# It calls _check_lambda_bound(lambdas, 0, np.inf)
# _check_lambda_bound converts inputs to atleast_1d arrays and checks bounds
# The function then returns 0.5 * lambdas ** 2

# Valid arguments for cgf_gaussian
ARGS = {
    'lambdas': np.array([0.5, 1.0, 1.5, 2.0]),
}
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 (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
compute_fder Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/diagnostics.py
code served (permissive licence) · get_code("82016b78b19009a1")
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

# Create small numpy arrays for the confidence interval bounds and true parameters
np.random.seed(0)

# Lower confidence bounds: shape (5,)
lcbs = np.array([-0.5, 0.2, 1.5, -0.1, 0.8])

# Upper confidence bounds: shape (5,)
ucbs = np.array([0.3, 0.8, 2.0, 0.5, 1.2])

# True parameters: shape (5,)
true_params = np.array([0.1, -0.3, 0.5, -0.2, 0.9])

ARGS = {
    "lcbs": lcbs,
    "ucbs": ucbs,
    "true_params": true_params
}
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
compute_miscoverage Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/diagnostics.py
code served (permissive licence) · get_code("cc828a98179b999e")
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 arguments for compute_miscoverage
# The function expects:
# - lcbs: lower confidence bounds (array-like)
# - ucbs: upper confidence bounds (array-like)
# - true_params: true parameter values (array-like)
# All should be sequences that can be compared element-wise

# Create small arrays with distinct sizes for clarity
np.random.seed(0)

# Generate sequences of confidence intervals and true parameters
# Using size 5 for the time sequence dimension
n_steps = 5

# True parameters - a sequence of 5 values
true_params = np.array([0.5, 0.6, 0.55, 0.65, 0.58])

# Lower confidence bounds - slightly below true params
lcbs = np.array([0.3, 0.4, 0.35, 0.45, 0.38])

# Upper confidence bounds - slightly above true params
ucbs = np.array([0.7, 0.8, 0.75, 0.85, 0.78])

ARGS = {
    "lcbs": lcbs,
    "ucbs": ucbs,
    "true_params": true_params
}
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
forecast_oracle Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/forecasters.py
code served (permissive licence) · get_code("6ae29d2768cb802e")
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 (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
from numpy.typing import ArrayLike

# Create a simple array-like input
ys = np.array([0, 1, 1, 0], dtype=np.float32)

# Define slack parameter
slack = 0.2

# Construct ARGS dict with all parameters
ARGS = {
    "ys": ys,
    "slack": slack,
}
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 (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
gamma_exponential_log_mixture Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/eprocess.py
code served (permissive licence) · get_code("227333666623e050")
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 (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

# Set random seeds for reproducibility
np.random.seed(0)

# Create small numpy arrays for sums and vs parameters
sums = np.array([0.5, -0.3, 1.2, -0.8])
vs = np.array([0.1, 0.2, 0.15, 0.25])

# Define scalar parameters
rho = 0.5
c = 0.8

# Construct the ARGS dictionary
ARGS = {
    "sums": sums,
    "vs": vs,
    "rho": rho,
    "c": c
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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_cgf Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/cgfs.py
code served (permissive licence) · get_code("b1208592ed448454")
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 function. 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 numpy as np

ARGS = {
    "name": "gaussian"
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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_color_by_index Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/plotting.py
code served (permissive licence) · get_code("ed3ad7621dd31dbb")
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 seaborn as sns

DEFAULT_PALETTE = 'colorblind'

def get_color_by_index(index: int, palette: str=DEFAULT_PALETTE):
    """Get color by integer indices."""
    colors = sns.color_palette(palette)
    return colors[index % len(colors)]

# Setup code
ARGS = {
    'index': 5,
    'palette': 'colorblind'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
seaborn — 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
get_colors Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/plotting.py
code served (permissive licence) · get_code("375a4934fe59f491")
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 _ColorPalette. 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 seaborn as sns

DEFAULT_PALETTE = 'colorblind'

ARGS = {
    'palette': 'colorblind'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
seaborn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type _ColorPalette — a _ColorPalette output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _ColorPalette 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
linear_kernel Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/kernels.py
code served (permissive licence) · get_code("1e2c7a761c21b61f")
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

# Create small numpy arrays for x and y
x = np.array([1.0, 2.0, 3.0, 4.0])
y = np.array([5.0, 6.0, 7.0, 8.0])

ARGS = {
    'x': x,
    'y': y,
}
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
shift_forecasts Ran
ran on a synthesized input
yjchoe/ComparingForecasters/comparecast/forecasters.py
code served (permissive licence) · get_code("f2d82687b97fd835")
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
from numpy.typing import ArrayLike

# Create a small forecast array
# The function expects forecasts (p_2, ..., p_{T+1}) made from data (y_1, ..., y_T)
# We'll create a simple array of 5 forecast values
ps = np.array([0.1, 0.2, 0.3, 0.4, 0.5])

# Define initial value for p_1
initial_value = 0.5

ARGS = {
    "ps": ps,
    "initial_value": initial_value
}
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
get_kernel Not yet run
unverified (harvested, no recorded run)
yjchoe/ComparingForecasters/comparecast/kernels.py
code served (permissive licence) · get_code("41805c814ad2afa2")
How it ran

no invocation recorded.

polynomial_kernel Not yet run
unverified (harvested, no recorded run)
yjchoe/ComparingForecasters/comparecast/kernels.py
code served (permissive licence) · get_code("86745abdc0db3187")
How it ran

no invocation recorded.

set_theme Not yet run
unverified (harvested, no recorded run)
yjchoe/ComparingForecasters/comparecast/plotting.py
code served (permissive licence) · get_code("34822df0396c065b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 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 15 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

Safe anytime-valid inference (SAVI) provides measures of statistical evidence and certainty -- e-processes for testing and confidence sequences for estimation -- that remain valid at all stopping times, accommodating continuous monitoring and analysis of accumulating data and optional stopping or continuation for any reason. These measures crucially rely on test martingales, which are nonnegative martingales starting at one. Since a test martingale is the wealth process of a player in a betting game, SAVI centrally employs game-theoretic intuition, language and mathematics. We summarize the SAVI goals and philosophy, and report recent advances in testing composite hypotheses and estimating functionals in nonparametric settings.

For agents

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

get_harvested_code_for_paper("2210.01948")
get_code_for_paper("2210.01948")
have("2210.01948")

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

curl https://syntology.ai/api/ran/2210.01948.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.01948.svg)](https://syntology.ai/paper/2210.01948)

Connect an agent — have() is free.