# A Deeper Look at Dataset Bias

arXiv: https://arxiv.org/abs/1505.01257 · HTML: https://mlsanity.com/paper/1505.01257/

## Key facts

- Cited by 87 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 May 2015.

## Built on by (most cited first, 30 of 87)

- [Kubric: A scalable dataset generator](https://mlsanity.com/paper/2203.03570/) (2022)
- [EnD: Entangling and Disentangling deep representations for bias correction](https://mlsanity.com/paper/2103.02023/) (2021)
- [A Survey on Bias in Visual Datasets](https://mlsanity.com/paper/2107.07919/) (2021)
- [Explainability of deep vision-based autonomous driving systems: Review and challenges](https://mlsanity.com/paper/2101.05307/) (2021)
- [A Prototype-Oriented Framework for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2110.12024/) (2021)
- [Understanding and Mitigating Annotation Bias in Facial Expression Recognition](https://mlsanity.com/paper/2108.08504/) (2021)
- [A Decade's Battle on Dataset Bias: Are We There Yet?](https://mlsanity.com/paper/2403.08632/) (2024)
- [Unbiased Supervised Contrastive Learning](https://mlsanity.com/paper/2211.05568/) (2022)
- [SelfHAR: Improving Human Activity Recognition through Self-training with Unlabeled Data](https://mlsanity.com/paper/2102.06073/) (2021)
- [Unsupervised Monocular Depth Estimation in Highly Complex Environments](https://mlsanity.com/paper/2107.13137/) (2021)
- [Domain-Specific Suppression for Adaptive Object Detection](https://mlsanity.com/paper/2105.03570/) (2021)
- [A Survey on Open Set Recognition](https://mlsanity.com/paper/2109.00893/) (2021)
- [Training Unbiased Diffusion Models From Biased Dataset](https://mlsanity.com/paper/2403.01189/) (2024)
- [Open-World Semantic Segmentation Including Class Similarity](https://mlsanity.com/paper/2403.07532/) (2024)
- [Neuro-Symbolic Hierarchical Rule Induction](https://mlsanity.com/paper/2112.13418/) (2021)
- [On the Cross-dataset Generalization in License Plate Recognition](https://mlsanity.com/paper/2201.00267/) (2022)
- [ColloSSL: Collaborative Self-Supervised Learning for Human Activity Recognition](https://mlsanity.com/paper/2202.00758/) (2022)
- [A First Look at Dataset Bias in License Plate Recognition](https://mlsanity.com/paper/2208.10657/) (2022)
- [Understanding Bias in Large-Scale Visual Datasets](https://mlsanity.com/paper/2412.01876/) (2024)
- [Do We Train on Test Data? The Impact of Near-Duplicates on License Plate Recognition](https://mlsanity.com/paper/2304.04653/) (2023)
- [CLIP-QDA: An Explainable Concept Bottleneck Model](https://mlsanity.com/paper/2312.00110/) (2023)
- [IFBiD: Inference-Free Bias Detection](https://mlsanity.com/paper/2109.04374/) (2021)
- [Accessing Vision Foundation Models via ImageNet-1K](https://mlsanity.com/paper/2407.10366/) (2024)
- [An Information Retrieval Approach to Building Datasets for Hate Speech Detection](https://mlsanity.com/paper/2106.09775/) (2021)
- [Analysis of Training Object Detection Models with Synthetic Data](https://mlsanity.com/paper/2211.16066/) (2022)
- [Unlearnable Examples Detection via Iterative Filtering](https://mlsanity.com/paper/2408.08143/) (2024)
- [Efficient Bias Mitigation Without Privileged Information](https://mlsanity.com/paper/2409.17691/) (2024)
- [Detecting discriminatory risk through data annotation based on Bayesian inferences](https://mlsanity.com/paper/2101.11358/) (2021)
- [A General Framework for Learning from Weak Supervision](https://mlsanity.com/paper/2402.01922/) (2024)
- [Exploiting Image Translations via Ensemble Self-Supervised Learning for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2107.06235/) (2021)

## Abstract (arXiv)

The presence of a bias in each image data collection has recently attracted a lot of attention in the computer vision community showing the limits in generalization of any learning method trained on a specific dataset. At the same time, with the rapid development of deep learning architectures, the activation values of Convolutional Neural Networks (CNN) are emerging as reliable and robust image descriptors. In this paper we propose to verify the potential of the DeCAF features when facing the dataset bias problem. We conduct a series of analyses looking at how existing datasets differ among each other and verifying the performance of existing debiasing methods under different representations. We learn important lessons on which part of the dataset bias problem can be considered solved and which open questions still need to be tackled.