SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.14434 · 2022

Flexible risk design using bi-directional dispersion

arXiv · PDF · Open in the Atlas

Code that ran

4 of 11 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 4 have no contract verdict; 0 failures of our instrument; 7 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/2203.14434.json · badge: /api/ran/2203.14434.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
feedbackward/bdd canonical (classifier label)
classifier basis not recorded on the link
4 ran · 7 no recorded run
FunctionStatusWhere it lives
eta_setter_barron Ran
ran on a synthesized input
feedbackward/bdd/bdd/setup_rescalers.py
code served (permissive licence) · get_code("22cf372a5f97f7c0")
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

# Test case 1: alpha < 2.0, sigma < threshold, interpolate=True
sigma_1 = 0.5
alpha_1 = 1.5
interpolate_1 = True
oce_flag_1 = False

# Test case 2: alpha > 2.0, sigma >= threshold
sigma_2 = 2.0
alpha_2 = 3.0
interpolate_2 = False
oce_flag_2 = False

# Test case 3: alpha = 2.0 (returns 1.0 directly)
sigma_3 = 1.5
alpha_3 = 2.0
interpolate_3 = True
oce_flag_3 = False

# Test case 4: alpha > 1.0, oce_flag=True, sigma < threshold
sigma_4 = 0.3
alpha_4 = 1.5
interpolate_4 = False
oce_flag_4 = True

ARGS = {
    "sigma": sigma_1,
    "alpha": alpha_1,
    "interpolate": interpolate_1,
    "oce_flag": oce_flag_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 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
eta_setter_barron_modulated Ran
ran on a synthesized input
feedbackward/bdd/bdd/setup_rescalers.py
code served (permissive licence) · get_code("919c742094330578")
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 float, 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

# The function takes 4 parameters: sigma, alpha, interpolate, oce_flag
# Let's analyze what values are valid based on the function's logic:
# - sigma: must be positive and finite (0 < sigma < inf)
# - alpha: can be any value, but certain combinations with other params are restricted
# - interpolate: boolean
# - oce_flag: boolean

# Test case 1: Simple case with sigma < 1.0, interpolate=False, oce_flag=False
ARGS = {
    "sigma": 0.5,
    "alpha": 2.0,
    "interpolate": False,
    "oce_flag": 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 float, 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
eta_setter_holland Ran
ran on a synthesized input
feedbackward/bdd/bdd/setup_rescalers.py
code served (permissive licence) · get_code("8cd33d2a55644d06")
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 float, 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

ARGS = {
    "sigma": 0.5,
    "interpolate": True,
    "oce_flag": 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 float, 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
parse_dro Ran
ran on a synthesized input
feedbackward/bdd/bdd/setup_losses.py
code served (permissive licence) · get_code("59c89ce2db15322d")
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 torch
import numpy as np

def parse_dro(atilde):
    shape = 2.0
    bound = (1.0 / (1.0 - atilde) - 1.0) ** 2.0 / 2.0
    return (bound, shape)

# atilde is used in arithmetic operations: (1.0 / (1.0 - atilde) - 1.0) ** 2.0 / 2.0
# It should be a scalar or numeric value that can be subtracted from 1.0 and used in division
# Using a float value less than 1.0 to avoid division by zero

ARGS = {
    "atilde": 0.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
d1_barron Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_dispersions.py
code served (permissive licence) · get_code("51a5f0778ca6c637")
How it ran

no invocation recorded.

d2_barron Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_dispersions.py
code served (permissive licence) · get_code("f52073ded31cc730")
How it ran

no invocation recorded.

dispersion_barron Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_dispersions.py
code served (permissive licence) · get_code("b214b4e5b4f76d92")
How it ran

no invocation recorded.

eval_model Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_eval.py
code served (permissive licence) · get_code("c481ddacab52025c")
How it ran

no invocation recorded.

eval_models Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_eval.py
code served (permissive licence) · get_code("02eda0fab364b714")
How it ran

no invocation recorded.

export_legend Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_results.py
code served (permissive licence) · get_code("30bbb8e87b49a63f")
How it ran

no invocation recorded.

train_epoch Not yet run
unverified (harvested, no recorded run)
feedbackward/bdd/bdd/setup_train.py
code served (permissive licence) · get_code("a1f40e7b727595d8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 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 11 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

Many novel notions of "risk" (e.g., CVaR, tilted risk, DRO risk) have been proposed and studied, but these risks are all at least as sensitive as the mean to loss tails on the upside, and tend to ignore deviations on the downside. We study a complementary new risk class that penalizes loss deviations in a bi-directional manner, while having more flexibility in terms of tail sensitivity than is offered by mean-variance. This class lets us derive high-probability learning guarantees without explicit gradient clipping, and empirical tests using both simulated and real data illustrate a high degree of control over key properties of the test loss distribution incurred by gradient-based learners.

For agents

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

get_harvested_code_for_paper("2203.14434")
get_code_for_paper("2203.14434")
have("2203.14434")

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

curl https://syntology.ai/api/ran/2203.14434.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.14434.svg)](https://syntology.ai/paper/2203.14434)

Connect an agent — have() is free.