SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.12594 · ICML · 2021

Directional Bias Amplification

Angelina Wang, Olga Russakovsky

arXiv · PDF · Open in the Atlas

Code that ran

1 of 3 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 0 have 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/2102.12594.json · badge: /api/ran/2102.12594.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
princetonvisualai/directional-bias-amp canonical (classifier label)
classifier basis: classifier_2025
1 ran · 2 no recorded run
FunctionStatusWhere it lives
biasamp_task_to_attribute Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
princetonvisualai/directional-bias-amp/directional_biasamp.py
code served (permissive licence) · get_code("59fb65ca1761eea2")
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 float64, 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

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

# Define dimensions
num_samples_test = 5
num_samples_train = 6
num_tasks = 3
num_attributes = 4

# Create task_labels: n x |T| (test set)
# Binary labels where 1 indicates the task is present
task_labels = np.array([
    [1, 0, 1],
    [0, 1, 0],
    [1, 1, 0],
    [0, 0, 1],
    [1, 0, 0]
], dtype=np.float32)

# Create attribute_labels: n x |A| (test set)
# Binary labels where 1 indicates the attribute is present
attribute_labels = np.array([
    [1, 0, 1, 0],
    [0, 1, 0, 1],
    [1, 1, 0, 0],
    [0, 0, 1, 1],
    [1, 0, 0, 1]
], dtype=np.float32)

# Create attribute_preds: n x |A| (test set predictions)
# Continuous predictions (not necessarily binary)
attribute_preds = np.array([
    [0.9, 0.1, 0.8, 0.2],
    [0.1, 0.95, 0.15, 0.85],
    [0.85, 0.8, 0.2, 0.1],
    [0.2, 0.1, 0.9, 0.88],
    [0.92, 0.05, 0.1, 0.95]
], dtype=np.float32)

# Create task_labels_train: m x |T| (train set)
task_labels_train = np.array([
    [1, 0, 1],
    [0, 1, 0],
    [1, 1, 0],
    [0, 0, 1],
    [1, 0, 0],
    [0, 1, 1]
], dtype=np.float32)

# Create attribute_labels_train: m x |A| (train set)
attribute_labels_train = np.array([
    [1, 0, 1, 0],
    [0, 1, 0, 1],
    [1, 1, 0, 0],
    [0, 0, 1, 1],
    [1, 0, 0, 1],
    [0, 1, 1, 0]
], dtype=np.float32)

# Create names
task_names = ['task_0', 'task_1', 'task_2']
attribute_names = ['attr_0', 'attr_1', 'attr_2', 'attr_3']
names = [task_names, attribute_names]

ARGS = {
    'task_labels': task_labels,
    'attribute_labels': attribute_labels,
    'attribute_preds': attribute_preds,
    'task_labels_train': task_labels_train,
    'attribute_labels_train': attribute_labels_train,
    'names': names
}
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 float64, 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
biasamp_attribute_to_task Not yet run
unverified (harvested, no recorded run)
princetonvisualai/directional-bias-amp/directional_biasamp.py
code served (permissive licence) · get_code("1d33232a5316d0ae")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

biasamp_task_to_attribute Not yet run
unverified (harvested, no recorded run)
princetonvisualai/directional-bias-amp/directional_biasamp.py
code served (permissive licence) · get_code("d3db227ca5b81e15")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 2 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 2 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

Mitigating bias in machine learning systems requires refining our understanding of bias propagation pathways: from societal structures to large-scale data to trained models to impact on society. In this work, we focus on one aspect of the problem, namely bias amplification: the tendency of models to amplify the biases present in the data they are trained on. A metric for measuring bias amplification was introduced in the seminal work by Zhao et al. (2017); however, as we demonstrate, this metric suffers from a number of shortcomings including conflating different types of bias amplification and failing to account for varying base rates of protected attributes. We introduce and analyze a new, decoupled metric for measuring bias amplification, BiasAmp → (Directional Bias Amplification). We thoroughly analyze and discuss both the technical assumptions and normative implications of this metric. We provide suggestions about its measurement by cautioning against predicting sensitive attributes, encouraging the use of confidence intervals due to fluctuations in the fairness of models across runs, and discussing the limitations of what this metric captures. Throughout this paper, we work to provide an interrogative look at the technical measurement of bias amplification, guided by our normative ideas of what we want it to encompass. Code is located at https: //github.com/princetonvisualai/ directional-bias-amp.

For agents

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

get_harvested_code_for_paper("2102.12594")
get_code_for_paper("2102.12594")
have("2102.12594")

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

curl https://syntology.ai/api/ran/2102.12594.json

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

[![Syntology run record](https://syntology.ai/api/ran/2102.12594.svg)](https://syntology.ai/paper/2102.12594)

Connect an agent — have() is free.