Evaluating Metrics for Bias in Word Embeddings
- URL: http://arxiv.org/abs/2111.07864v1
- Date: Mon, 15 Nov 2021 16:07:15 GMT
- Title: Evaluating Metrics for Bias in Word Embeddings
- Authors: Sarah Schr\"oder, Alexander Schulz, Philip Kenneweg, Robert Feldhans,
Fabian Hinder and Barbara Hammer
- Abstract summary: We formalize a bias definition based on the ideas from previous works and derive conditions for bias metrics.
We propose a new metric, SAME, to address the shortcomings of existing metrics and mathematically prove that SAME behaves appropriately.
- Score: 64.55554083622258
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Over the last years, word and sentence embeddings have established as text
preprocessing for all kinds of NLP tasks and improved the performances
significantly. Unfortunately, it has also been shown that these embeddings
inherit various kinds of biases from the training data and thereby pass on
biases present in society to NLP solutions. Many papers attempted to quantify
bias in word or sentence embeddings to evaluate debiasing methods or compare
different embedding models, usually with cosine-based metrics. However, lately
some works have raised doubts about these metrics showing that even though such
metrics report low biases, other tests still show biases. In fact, there is a
great variety of bias metrics or tests proposed in the literature without any
consensus on the optimal solutions. Yet we lack works that evaluate bias
metrics on a theoretical level or elaborate the advantages and disadvantages of
different bias metrics. In this work, we will explore different cosine based
bias metrics. We formalize a bias definition based on the ideas from previous
works and derive conditions for bias metrics. Furthermore, we thoroughly
investigate the existing cosine-based metrics and their limitations to show why
these metrics can fail to report biases in some cases. Finally, we propose a
new metric, SAME, to address the shortcomings of existing metrics and
mathematically prove that SAME behaves appropriately.
Related papers
- Cobra Effect in Reference-Free Image Captioning Metrics [58.438648377314436]
A proliferation of reference-free methods, leveraging visual-language pre-trained models (VLMs), has emerged.
In this paper, we study if there are any deficiencies in reference-free metrics.
We employ GPT-4V as an evaluative tool to assess generated sentences and the result reveals that our approach achieves state-of-the-art (SOTA) performance.
arXiv Detail & Related papers (2024-02-18T12:36:23Z) - Semantic Properties of cosine based bias scores for word embeddings [52.13994416317707]
We propose requirements for bias scores to be considered meaningful for quantifying biases.
We analyze cosine based scores from the literature with regard to these requirements.
We underline these findings with experiments to show that the bias scores' limitations have an impact in the application case.
arXiv Detail & Related papers (2024-01-27T20:31:10Z) - A Bayesian approach to uncertainty in word embedding bias estimation [0.0]
Multiple measures, such as WEAT or MAC, attempt to quantify the magnitude of bias present in word embeddings in terms of a single-number metric.
We show that similar results can be easily obtained using such methods even if the data are generated by a null model lacking the intended bias.
We propose a Bayesian alternative: hierarchical Bayesian modeling, which enables a more uncertainty-sensitive inspection of bias in word embeddings.
arXiv Detail & Related papers (2023-06-15T11:48:50Z) - Testing Occupational Gender Bias in Language Models: Towards Robust Measurement and Zero-Shot Debiasing [98.07536837448293]
Large language models (LLMs) have been shown to exhibit a variety of harmful, human-like biases against various demographics.
We introduce a list of desiderata for robustly measuring biases in generative language models.
We then use this benchmark to test several state-of-the-art open-source LLMs, including Llama, Mistral, and their instruction-tuned versions.
arXiv Detail & Related papers (2022-12-20T22:41:24Z) - Choose Your Lenses: Flaws in Gender Bias Evaluation [29.16221451643288]
We assess the current paradigm of gender bias evaluation and identify several flaws in it.
First, we highlight the importance of extrinsic bias metrics that measure how a model's performance on some task is affected by gender.
Second, we find that datasets and metrics are often coupled, and discuss how their coupling hinders the ability to obtain reliable conclusions.
arXiv Detail & Related papers (2022-10-20T17:59:55Z) - The SAME score: Improved cosine based bias score for word embeddings [63.24247894974291]
We provide a bias definition based on the ideas from the literature and derive novel requirements for bias scores.
We propose a new bias score, SAME, to address the shortcomings of existing bias scores and show empirically that SAME is better suited to quantify biases in word embeddings.
arXiv Detail & Related papers (2022-03-28T09:28:13Z) - Information-Theoretic Bias Reduction via Causal View of Spurious
Correlation [71.9123886505321]
We propose an information-theoretic bias measurement technique through a causal interpretation of spurious correlation.
We present a novel debiasing framework against the algorithmic bias, which incorporates a bias regularization loss.
The proposed bias measurement and debiasing approaches are validated in diverse realistic scenarios.
arXiv Detail & Related papers (2022-01-10T01:19:31Z) - Measuring Fairness with Biased Rulers: A Survey on Quantifying Biases in
Pretrained Language Models [2.567384209291337]
An increasing awareness of biased patterns in natural language processing resources has motivated many metrics to quantify bias' and fairness'
We survey the existing literature on fairness metrics for pretrained language models and experimentally evaluate compatibility.
We find that many metrics are not compatible and highly depend on (i) templates, (ii) attribute and target seeds and (iii) the choice of embeddings.
arXiv Detail & Related papers (2021-12-14T15:04:56Z) - Intrinsic Bias Metrics Do Not Correlate with Application Bias [12.588713044749179]
This research examines whether easy-to-measure intrinsic metrics correlate well to real world extrinsic metrics.
We measure both intrinsic and extrinsic bias across hundreds of trained models covering different tasks and experimental conditions.
We advise that efforts to debias embedding spaces be always also paired with measurement of downstream model bias, and suggest that that community increase effort into making downstream measurement more feasible via creation of additional challenge sets and annotated test data.
arXiv Detail & Related papers (2020-12-31T18:59:44Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.