Correcting Annotator Bias in Training Data: Population-Aligned Instance Replication (PAIR)

• URL: http://arxiv.org/abs/2501.06826v1
• Date: Sun, 12 Jan 2025 14:39:26 GMT
• Title: Correcting Annotator Bias in Training Data: Population-Aligned Instance Replication (PAIR)
• Authors: Stephanie Eckman, Bolei Ma, Christoph Kern, Rob Chew, Barbara Plank, Frauke Kreuter,
• Abstract summary: Models trained on crowdsourced labels may not reflect broader population views when annotator pools are not representative.<n>We propose Population-Aligned Instance Replication (PAIR), a method to address this bias through statistical adjustment.
• Score: 24.280324949484406
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Models trained on crowdsourced labels may not reflect broader population views when annotator pools are not representative. Since collecting representative labels is challenging, we propose Population-Aligned Instance Replication (PAIR), a method to address this bias through statistical adjustment. Using a simulation study of hate speech and offensive language detection, we create two types of annotators with different labeling tendencies and generate datasets with varying proportions of the types. Models trained on unbalanced annotator pools show poor calibration compared to those trained on representative data. However, PAIR, which duplicates labels from underrepresented annotator groups to match population proportions, significantly reduces bias without requiring new data collection. These results suggest statistical techniques from survey research can help align model training with target populations even when representative annotator pools are unavailable. We conclude with three practical recommendations for improving training data quality.

Related papers

• No Training Wheels: Steering Vectors for Bias Correction at Inference Time [0.0]
  We propose a cheap, training-free method inspired by steering vectors used to edit behaviors in large language models.<n>We compute the difference in mean activations between majority and minority groups to define a "bias vector"<n>This leads to reduced classification bias and improved worst-group accuracy.
  arXiv  Detail & Related papers  (2025-06-23T12:58:54Z)
• Active Data Sampling and Generation for Bias Remediation [0.0]
  A mixed active sampling and data generation strategy -- called samplation -- is proposed to compensate during fine-tuning of a pre-trained classifer the unfair classifications it produces. Using as case study Deep Models for visual semantic role labeling, the proposed method has been able to fully cure a simulated gender bias starting from a 90/10 imbalance.
  arXiv  Detail & Related papers  (2025-03-26T10:42:15Z)
• Dataset Representativeness and Downstream Task Fairness [24.570493924073524]
  We demonstrate that there is a natural tension between dataset representativeness and group-fairness of classifiers trained on that dataset. We also find that over-sampling underrepresented groups can result in classifiers which exhibit greater bias to those groups.
  arXiv  Detail & Related papers  (2024-06-28T18:11:16Z)
• Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selection [80.85902083005237]
  We introduce Data Debiasing with Datamodels (D3M), a debiasing approach which isolates and removes specific training examples that drive the model's failures on minority groups.
  arXiv  Detail & Related papers  (2024-06-24T17:51:01Z)
• Unsupervised Concept Discovery Mitigates Spurious Correlations [45.48778210340187]
  Models prone to spurious correlations in training data often produce brittle predictions and introduce unintended biases. In this paper, we establish a novel connection between unsupervised object-centric learning and mitigation of spurious correlations. We introduce CoBalT: a concept balancing technique that effectively mitigates spurious correlations without requiring human labeling of subgroups.
  arXiv  Detail & Related papers  (2024-02-20T20:48:00Z)
• Debiased Sample Selection for Combating Noisy Labels [24.296451733127956]
  We propose a noIse-Tolerant Expert Model (ITEM) for debiased learning in sample selection. Specifically, to mitigate the training bias, we design a robust network architecture that integrates with multiple experts. By training on the mixture of two class-discriminative mini-batches, the model mitigates the effect of the imbalanced training set.
  arXiv  Detail & Related papers  (2024-01-24T10:37:28Z)
• IBADR: an Iterative Bias-Aware Dataset Refinement Framework for Debiasing NLU models [52.03761198830643]
  We propose IBADR, an Iterative Bias-Aware dataset Refinement framework. We first train a shallow model to quantify the bias degree of samples in the pool. Then, we pair each sample with a bias indicator representing its bias degree, and use these extended samples to train a sample generator. In this way, this generator can effectively learn the correspondence relationship between bias indicators and samples.
  arXiv  Detail & Related papers  (2023-11-01T04:50:38Z)
• Tackling Diverse Minorities in Imbalanced Classification [80.78227787608714]
  Imbalanced datasets are commonly observed in various real-world applications, presenting significant challenges in training classifiers. We propose generating synthetic samples iteratively by mixing data samples from both minority and majority classes. We demonstrate the effectiveness of our proposed framework through extensive experiments conducted on seven publicly available benchmark datasets.
  arXiv  Detail & Related papers  (2023-08-28T18:48:34Z)
• Stubborn Lexical Bias in Data and Models [50.79738900885665]
  We use a new statistical method to examine whether spurious patterns in data appear in models trained on the data. We apply an optimization approach to *reweight* the training data, reducing thousands of spurious correlations. Surprisingly, though this method can successfully reduce lexical biases in the training data, we still find strong evidence of corresponding bias in the trained models.
  arXiv  Detail & Related papers  (2023-06-03T20:12:27Z)
• Improving Heterogeneous Model Reuse by Density Estimation [105.97036205113258]
  This paper studies multiparty learning, aiming to learn a model using the private data of different participants. Model reuse is a promising solution for multiparty learning, assuming that a local model has been trained for each party.
  arXiv  Detail & Related papers  (2023-05-23T09:46:54Z)
• Non-Invasive Fairness in Learning through the Lens of Data Drift [88.37640805363317]
  We show how to improve the fairness of Machine Learning models without altering the data or the learning algorithm. We use a simple but key insight: the divergence of trends between different populations, and, consecutively, between a learned model and minority populations, is analogous to data drift. We explore two strategies (model-splitting and reweighing) to resolve this drift, aiming to improve the overall conformance of models to the underlying data.
  arXiv  Detail & Related papers  (2023-03-30T17:30:42Z)
• 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)
• Consistent Range Approximation for Fair Predictive Modeling [10.613912061919775]
  The framework builds predictive models that are certifiably fair on the target population, regardless of the availability of external data during training. The framework's efficacy is demonstrated through evaluations on real data, showing substantial improvement over existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-12-21T08:27:49Z)
• Toward Annotator Group Bias in Crowdsourcing [26.754873038110595]
  We show that annotators within the same demographic group tend to show consistent group bias in annotation tasks. We develop a novel probabilistic graphical framework GroupAnno to capture annotator group bias with a new extended Expectation Maximization (EM) training algorithm.
  arXiv  Detail & Related papers  (2021-10-08T05:54:36Z)
• 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)
• Neighborhood-based Pooling for Population-level Label Distribution Learning [5.790608871289107]
  Supervised machine learning often requires human-annotated data. Population-level label distribution learning treats the collection of annotations for each data item as a sample of the opinions of a population of human annotators. We propose an algorithmic framework and new statistical tests for PLDL that account for sampling size.
  arXiv  Detail & Related papers  (2020-03-16T18:52:56Z)

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.