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

arXiv: https://arxiv.org/abs/2010.15052 · HTML: https://mlsanity.com/paper/2010.15052/

## Key facts

- Cited by 107 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Oct 2020.

## Built on by (most cited first, 30 of 107)

- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Video Diffusion Models](https://mlsanity.com/paper/2204.03458/) (2022)
- [Vector-quantized Image Modeling with Improved VQGAN](https://mlsanity.com/paper/2110.04627/) (2021)
- [Muse: Text-To-Image Generation via Masked Generative Transformers](https://mlsanity.com/paper/2301.00704/) (2023)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models](https://mlsanity.com/paper/2202.04053/) (2022)
- [Fake it till you make it: Learning transferable representations from synthetic ImageNet clones](https://mlsanity.com/paper/2212.08420/) (2022)
- [Sociotechnical Safety Evaluation of Generative AI Systems](https://mlsanity.com/paper/2310.11986/) (2023)
- [Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?](https://mlsanity.com/paper/2202.06675/) (2022)
- [Fair Diffusion: Instructing Text-to-Image Generation Models on Fairness](https://mlsanity.com/paper/2302.10893/) (2023)
- [Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localization](https://mlsanity.com/paper/2205.07839/) (2022)
- [A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learning](https://mlsanity.com/paper/2203.11933/) (2022)
- [Are Gender-Neutral Queries Really Gender-Neutral? Mitigating Gender Bias in Image Search](https://mlsanity.com/paper/2109.05433/) (2021)
- [Extending the WILDS Benchmark for Unsupervised Adaptation](https://mlsanity.com/paper/2112.05090/) (2021)
- [A Survey on Bias in Visual Datasets](https://mlsanity.com/paper/2107.07919/) (2021)
- [Normalizing Flows for Human Pose Anomaly Detection](https://mlsanity.com/paper/2211.10946/) (2022)
- [Improved Transformer for High-Resolution GANs](https://mlsanity.com/paper/2106.07631/) (2021)
- [Use All The Labels: A Hierarchical Multi-Label Contrastive Learning Framework](https://mlsanity.com/paper/2204.13207/) (2022)
- [Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluation](https://mlsanity.com/paper/2205.04610/) (2022)
- [FACET: Fairness in Computer Vision Evaluation Benchmark](https://mlsanity.com/paper/2309.00035/) (2023)
- [PASS: An ImageNet replacement for self-supervised pretraining without humans](https://mlsanity.com/paper/2109.13228/) (2021)
- [Mitigating Dataset Harms Requires Stewardship: Lessons from 1000 Papers](https://mlsanity.com/paper/2108.02922/) (2021)
- [Inspecting the Geographical Representativeness of Images from Text-to-Image Models](https://mlsanity.com/paper/2305.11080/) (2023)
- [Contrastive Language-Vision AI Models Pretrained on Web-Scraped Multimodal Data Exhibit Sexual Objectification Bias](https://mlsanity.com/paper/2212.11261/) (2022)
- [A Survey on Intersectional Fairness in Machine Learning: Notions, Mitigation, and Challenges](https://mlsanity.com/paper/2305.06969/) (2023)
- [American == White in Multimodal Language-and-Image AI](https://mlsanity.com/paper/2207.00691/) (2022)
- [T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation](https://mlsanity.com/paper/2306.00905/) (2023)
- [Supervised Transfer Learning at Scale for Medical Imaging](https://mlsanity.com/paper/2101.05913/) (2021)
- [Overwriting Pretrained Bias with Finetuning Data](https://mlsanity.com/paper/2303.06167/) (2023)

## Abstract (arXiv)

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.