Debiasing Pre-trained Contextualised Embeddings

• URL: http://arxiv.org/abs/2101.09523v1
• Date: Sat, 23 Jan 2021 15:28:48 GMT
• Title: Debiasing Pre-trained Contextualised Embeddings
• Authors: Masahiro Kaneko and Danushka Bollegala
• Abstract summary: We propose a fine-tuning method that can be applied at token- or sentence-levels to debias pre-trained contextualised embeddings. Using gender bias as an illustrative example, we then conduct a systematic study using several state-of-the-art (SoTA) contextualised representations. We find that applying token-level debiasing for all tokens and across all layers of a contextualised embedding model produces the best performance.
• Score: 28.378270372391498
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In comparison to the numerous debiasing methods proposed for the static non-contextualised word embeddings, the discriminative biases in contextualised embeddings have received relatively little attention. We propose a fine-tuning method that can be applied at token- or sentence-levels to debias pre-trained contextualised embeddings. Our proposed method can be applied to any pre-trained contextualised embedding model, without requiring to retrain those models. Using gender bias as an illustrative example, we then conduct a systematic study using several state-of-the-art (SoTA) contextualised representations on multiple benchmark datasets to evaluate the level of biases encoded in different contextualised embeddings before and after debiasing using the proposed method. We find that applying token-level debiasing for all tokens and across all layers of a contextualised embedding model produces the best performance. Interestingly, we observe that there is a trade-off between creating an accurate vs. unbiased contextualised embedding model, and different contextualised embedding models respond differently to this trade-off.

Related papers

• Ensembling Finetuned Language Models for Text Classification [55.15643209328513]
  Finetuning is a common practice across different communities to adapt pretrained models to particular tasks. ensembles of neural networks are typically used to boost performance and provide reliable uncertainty estimates. We present a metadataset with predictions from five large finetuned models on six datasets and report results of different ensembling strategies.
  arXiv  Detail & Related papers  (2024-10-25T09:15:54Z)
• ConceptDrift: Uncovering Biases through the Lens of Foundational Models [5.025665239455297]
  We propose ConceptDrift, a method which analyzes the weights of a linear probe, learned on top a foundational model. We empirically prove the efficacy of our method, by significantly improving zero-shot performance with biased-augmented prompting.
  arXiv  Detail & Related papers  (2024-10-24T17:59:16Z)
• CosFairNet:A Parameter-Space based Approach for Bias Free Learning [1.9116784879310025]
  Deep neural networks trained on biased data often inadvertently learn unintended inference rules. We introduce a novel approach to address bias directly in the model's parameter space, preventing its propagation across layers. We show enhanced classification accuracy and debiasing effectiveness across various synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2024-10-19T13:06:40Z)
• Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical [66.57396042747706]
  Complementary-label learning is a weakly supervised learning problem. We propose a consistent approach that does not rely on the uniform distribution assumption. We find that complementary-label learning can be expressed as a set of negative-unlabeled binary classification problems.
  arXiv  Detail & Related papers  (2023-11-27T02:59:17Z)
• Fair Enough: Standardizing Evaluation and Model Selection for Fairness Research in NLP [64.45845091719002]
  Modern NLP systems exhibit a range of biases, which a growing literature on model debiasing attempts to correct. This paper seeks to clarify the current situation and plot a course for meaningful progress in fair learning.
  arXiv  Detail & Related papers  (2023-02-11T14:54:00Z)
• Debiasing Vision-Language Models via Biased Prompts [79.04467131711775]
  We propose a general approach for debiasing vision-language foundation models by projecting out biased directions in the text embedding. We show that debiasing only the text embedding with a calibrated projection matrix suffices to yield robust classifiers and fair generative models.
  arXiv  Detail & Related papers  (2023-01-31T20:09:33Z)
• BiasBed -- Rigorous Texture Bias Evaluation [21.55506905780658]
  We introduce BiasBed, a testbed for texture- and style-biased training. It comes with rigorous hypothesis testing to gauge the significance of the results. E.g., we find that some algorithms proposed in the literature do not significantly mitigate the impact of style bias at all.
  arXiv  Detail & Related papers  (2022-11-23T18:22:59Z)
• Mind Your Bias: A Critical Review of Bias Detection Methods for Contextual Language Models [2.170169149901781]
  We conduct a rigorous analysis and comparison of bias detection methods for contextual language models. Our results show that minor design and implementation decisions (or errors) have a substantial and often significant impact on the derived bias scores.
  arXiv  Detail & Related papers  (2022-11-15T19:27:54Z)
• 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)
• Learning Debiased Models with Dynamic Gradient Alignment and Bias-conflicting Sample Mining [39.00256193731365]
  Deep neural networks notoriously suffer from dataset biases which are detrimental to model robustness, generalization and fairness. We propose a two-stage debiasing scheme to combat against the intractable unknown biases.
  arXiv  Detail & Related papers  (2021-11-25T14:50:10Z)
• Balancing out Bias: Achieving Fairness Through Training Reweighting [58.201275105195485]
  Bias in natural language processing arises from models learning characteristics of the author such as gender and race. Existing methods for mitigating and measuring bias do not directly account for correlations between author demographics and linguistic variables. This paper introduces a very simple but highly effective method for countering bias using instance reweighting.
  arXiv  Detail & Related papers  (2021-09-16T23:40:28Z)

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.