Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biases

• URL: http://arxiv.org/abs/2010.15052v3
• Date: Wed, 27 Jan 2021 18:48:10 GMT
• Title: Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biases
• Authors: Ryan Steed and Aylin Caliskan
• Abstract summary: We develop a novel method for quantifying biased associations between representations of social concepts and attributes in images. We find that state-of-the-art unsupervised models trained on ImageNet, a popular benchmark image dataset, automatically learn racial, gender, and intersectional biases.
• Score: 3.0349733976070015
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Recent advances in machine learning leverage massive datasets of unlabeled images from the web to learn general-purpose image representations for tasks from image classification to face recognition. But do unsupervised computer vision models automatically learn implicit patterns and embed social biases that could have harmful downstream effects? We develop a novel method for quantifying biased associations between representations of social concepts and attributes in images. We find that state-of-the-art unsupervised models trained on ImageNet, a popular benchmark image dataset curated from internet images, automatically learn racial, gender, and intersectional biases. We replicate 8 previously documented human biases from social psychology, from the innocuous, as with insects and flowers, to the potentially harmful, as with race and gender. Our results closely match three hypotheses about intersectional bias from social psychology. For the first time in unsupervised computer vision, we also quantify implicit human biases about weight, disabilities, and several ethnicities. When compared with statistical patterns in online image datasets, our findings suggest that machine learning models can automatically learn bias from the way people are stereotypically portrayed on the web.

Related papers

• When Does Perceptual Alignment Benefit Vision Representations? [76.32336818860965]
  We investigate how aligning vision model representations to human perceptual judgments impacts their usability. We find that aligning models to perceptual judgments yields representations that improve upon the original backbones across many downstream tasks. Our results suggest that injecting an inductive bias about human perceptual knowledge into vision models can contribute to better representations.
  arXiv  Detail & Related papers  (2024-10-14T17:59:58Z)
• Classes Are Not Equal: An Empirical Study on Image Recognition Fairness [100.36114135663836]
  We experimentally demonstrate that classes are not equal and the fairness issue is prevalent for image classification models across various datasets. Our findings reveal that models tend to exhibit greater prediction biases for classes that are more challenging to recognize. Data augmentation and representation learning algorithms improve overall performance by promoting fairness to some degree in image classification.
  arXiv  Detail & Related papers  (2024-02-28T07:54:50Z)
• Examining Pathological Bias in a Generative Adversarial Network Discriminator: A Case Study on a StyleGAN3 Model [0.31572294564491876]
  Generative adversarial networks (GANs) generate photorealistic faces that are often indistinguishable by humans from real faces. We find pathological internal color and luminance biases in the discriminator of a pre-trained StyleGAN3-r model. We also find that the discriminator systematically stratifies scores by both image- and face-level qualities and that this disproportionately affects images across gender, race, and other categories.
  arXiv  Detail & Related papers  (2024-02-15T08:34:21Z)
• Probing Intersectional Biases in Vision-Language Models with Counterfactual Examples [5.870913541790421]
  We employ text-to-image diffusion models to produce counterfactual examples for probing intserctional social biases at scale. Our approach utilizes Stable Diffusion with cross attention control to produce sets of counterfactual image-text pairs. We conduct extensive experiments using our generated dataset which reveal the intersectional social biases present in state-of-the-art VLMs.
  arXiv  Detail & Related papers  (2023-10-04T17:25:10Z)
• Gender Biases in Automatic Evaluation Metrics for Image Captioning [87.15170977240643]
  We conduct a systematic study of gender biases in model-based evaluation metrics for image captioning tasks. We demonstrate the negative consequences of using these biased metrics, including the inability to differentiate between biased and unbiased generations. We present a simple and effective way to mitigate the metric bias without hurting the correlations with human judgments.
  arXiv  Detail & Related papers  (2023-05-24T04:27:40Z)
• Unravelling the Effect of Image Distortions for Biased Prediction of Pre-trained Face Recognition Models [86.79402670904338]
  We evaluate the performance of four state-of-the-art deep face recognition models in the presence of image distortions. We have observed that image distortions have a relationship with the performance gap of the model across different subgroups.
  arXiv  Detail & Related papers  (2021-08-14T16:49:05Z)
• Learning from Failure: Training Debiased Classifier from Biased Classifier [76.52804102765931]
  We show that neural networks learn to rely on spurious correlation only when it is "easier" to learn than the desired knowledge. We propose a failure-based debiasing scheme by training a pair of neural networks simultaneously. Our method significantly improves the training of the network against various types of biases in both synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2020-07-06T07:20:29Z)
• Gender Slopes: Counterfactual Fairness for Computer Vision Models by Attribute Manipulation [4.784524967912113]
  Automated computer vision systems have been applied in many domains including security, law enforcement, and personal devices. Recent reports suggest that these systems may produce biased results, discriminating against people in certain demographic groups. We propose to use an encoder-decoder network developed for image manipulation to synthesize facial images varying in the dimensions of gender and race.
  arXiv  Detail & Related papers  (2020-05-21T02:33:28Z)
• InsideBias: Measuring Bias in Deep Networks and Application to Face Gender Biometrics [73.85525896663371]
  This work explores the biases in learning processes based on deep neural network architectures. We employ two gender detection models based on popular deep neural networks. We propose InsideBias, a novel method to detect biased models.
  arXiv  Detail & Related papers  (2020-04-14T15:20:50Z)
• A Set of Distinct Facial Traits Learned by Machines Is Not Predictive of Appearance Bias in the Wild [3.0349733976070015]
  We train a learning model on human subjects' first impressions of personality traits in other faces as measured by social psychologists. We find that features extracted with FaceNet can be used to predict human appearance bias scores for deliberately manipulated faces. In contrast to work with human biases in social psychology, the model does not find a significant signal correlating politicians' vote shares with perceived competence bias.
  arXiv  Detail & Related papers  (2020-02-13T17:09:27Z)

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.