# When does Bias Transfer in Transfer Learning?

arXiv: https://arxiv.org/abs/2207.02842 · HTML: https://mlsanity.com/paper/2207.02842/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 56 references that are in the map.
- First posted on arXiv 6 Jul 2022.

## Builds on (most cited first, 29 of 29)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain](https://mlsanity.com/paper/1708.06733/) (2017)
- [ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases](https://mlsanity.com/paper/1705.02315/) (2017)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [Exploring the Limits of Weakly Supervised Pretraining](https://mlsanity.com/paper/1805.00932/) (2018)
- [Do Better ImageNet Models Transfer Better?](https://mlsanity.com/paper/1805.08974/) (2018)
- [R-FCN: Object Detection via Region-based Fully Convolutional Networks](https://mlsanity.com/paper/1605.06409/) (2016)
- [Noise or Signal: The Role of Image Backgrounds in Object Recognition](https://mlsanity.com/paper/2006.09994/) (2020)
- [Do Adversarially Robust ImageNet Models Transfer Better?](https://mlsanity.com/paper/2007.08489/) (2020)
- [Certified Defenses for Data Poisoning Attacks](https://mlsanity.com/paper/1706.03691/) (2017)
- [Label-Consistent Backdoor Attacks](https://mlsanity.com/paper/1912.02771/) (2019)
- [Interactive Supercomputing on 40,000 Cores for Machine Learning and Data Analysis](https://mlsanity.com/paper/1807.07814/) (2018)
- [Don't Judge an Object by Its Context: Learning to Overcome Contextual Bias](https://mlsanity.com/paper/2001.03152/) (2020)
- [A Comprehensive Study on Face Recognition Biases Beyond Demographics](https://mlsanity.com/paper/2103.01592/) (2021)
- [CheXtransfer: Performance and Parameter Efficiency of ImageNet Models for Chest X-Ray Interpretation](https://mlsanity.com/paper/2101.06871/) (2021)
- [Object Recognition with and without Objects](https://mlsanity.com/paper/1611.06596/) (2016)
- [Transfer Learning from Deep Features for Remote Sensing and Poverty Mapping](https://mlsanity.com/paper/1510.00098/) (2015)
- [3DB: A Framework for Debugging Computer Vision Models](https://mlsanity.com/paper/2106.03805/) (2021)
- [Adversarially-Trained Deep Nets Transfer Better: Illustration on Image Classification](https://mlsanity.com/paper/2007.05869/) (2020)
- [Self-Driving Car Steering Angle Prediction Based on Image Recognition](https://mlsanity.com/paper/1912.05440/) (2019)

## Built on by (most cited first, 23 of 23)

- [Fake it till you make it: Learning transferable representations from synthetic ImageNet clones](https://mlsanity.com/paper/2212.08420/) (2022)
- [Overwriting Pretrained Bias with Finetuning Data](https://mlsanity.com/paper/2303.06167/) (2023)
- [ModelDiff: A Framework for Comparing Learning Algorithms](https://mlsanity.com/paper/2211.12491/) (2022)
- [Understanding the Detrimental Class-level Effects of Data Augmentation](https://mlsanity.com/paper/2401.01764/) (2023)
- [BiasPruner: Debiased Continual Learning for Medical Image Classification](https://mlsanity.com/paper/2407.08609/) (2024)
- [Bias in Pruned Vision Models: In-Depth Analysis and Countermeasures](https://mlsanity.com/paper/2304.12622/) (2023)
- [Bridging the Digital Divide: Performance Variation across Socio-Economic Factors in Vision-Language Models](https://mlsanity.com/paper/2311.05746/) (2023)
- [Meta-learning in healthcare: A survey](https://mlsanity.com/paper/2308.02877/) (2023)
- [Ask Your Distribution Shift if Pre-Training is Right for You](https://mlsanity.com/paper/2403.00194/) (2024)
- [AI Supply Chains: An Emerging Ecosystem of AI Actors, Products, and Services](https://mlsanity.com/paper/2504.20185/) (2025)
- [Astroformer: More Data Might not be all you need for Classification](https://mlsanity.com/paper/2304.05350/) (2023)
- [Annotations on a Budget: Leveraging Geo-Data Similarity to Balance Model Performance and Annotation Cost](https://mlsanity.com/paper/2403.07687/) (2024)
- [No Detail Left Behind: Revisiting Self-Retrieval for Fine-Grained Image Captioning](https://mlsanity.com/paper/2409.03025/) (2024)
- [ALBAR: Adversarial Learning approach to mitigate Biases in Action Recognition](https://mlsanity.com/paper/2502.00156/) (2025)
- [Variation of Gender Biases in Visual Recognition Models Before and After Finetuning](https://mlsanity.com/paper/2303.07615/) (2023)
- [Federated and Transfer Learning for Cancer Detection Based on Image Analysis](https://mlsanity.com/paper/2405.20126/) (2024)
- [Injecting Undetectable Backdoors in Obfuscated Neural Networks and Language Models](https://mlsanity.com/paper/2406.05660/) (2024)
- [TorchQL: A Programming Framework for Integrity Constraints in Machine Learning](https://mlsanity.com/paper/2308.06686/) (2023)
- [Views Can Be Deceiving: Improved SSL Through Feature Space Augmentation](https://mlsanity.com/paper/2406.18562/) (2024)
- [Continual Learning in the Presence of Spurious Correlation](https://mlsanity.com/paper/2303.11863/) (2023)
- [Pick your Poison: Undetectability versus Robustness in Data Poisoning Attacks](https://mlsanity.com/paper/2305.09671/) (2023)
- [Assessing Intersectional Bias in Representations of Pre-Trained Image Recognition Models](https://mlsanity.com/paper/2506.03664/) (2025)
- [Diagnosing Shortcut-Induced Rigidity in Continual Learning: The Einstellung Rigidity Index (ERI)](https://mlsanity.com/paper/2510.00475/) (2025)

## Abstract (arXiv)

Using transfer learning to adapt a pre-trained "source model" to a downstream "target task" can dramatically increase performance with seemingly no downside. In this work, we demonstrate that there can exist a downside after all: bias transfer, or the tendency for biases of the source model to persist even after adapting the model to the target class. Through a combination of synthetic and natural experiments, we show that bias transfer both (a) arises in realistic settings (such as when pre-training on ImageNet or other standard datasets) and (b) can occur even when the target dataset is explicitly de-biased. As transfer-learned models are increasingly deployed in the real world, our work highlights the importance of understanding the limitations of pre-trained source models. Code is available at https://github.com/MadryLab/bias-transfer