mlsanityData to 7 Oct 2026

IFBiD: Inference-Free Bias Detection

arXiv 2109.04374 ↗First posted 9 Sep 2021cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

This paper is the first to explore an automatic way to detect bias in deep convolutional neural networks by simply looking at their weights. Furthermore, it is also a step towards understanding neural networks and how they work. We show that it is indeed possible to know if a model is biased or not simply by looking at its weights, without the model inference for an specific input. We analyze how bias is encoded in the weights of deep networks through a toy example using the Colored MNIST database and we also provide a realistic case study in gender detection from face images using state-of-the-art methods and experimental resources. To do so, we generated two databases with 36K and 48K biased models each. In the MNIST models we were able to detect whether they presented a strong or low bias with more than 99% accuracy, and we were also able to classify between four levels of bias with more than 70% accuracy. For the face models, we achieved 90% accuracy in distinguishing between models biased towards Asian, Black, or Caucasian ethnicity.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 38 of its 57 references that are in the map. First posted on arXiv 9 Sep 2021.

Builds on

38 of 57 references.

References · 38

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  2. 2013Intriguing properties of neural networkscited by 5,114
  3. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  5. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  7. 2019Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AIcited by 1,543
  8. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  9. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  10. 2018Sanity Checks for Saliency Mapscited by 1,149
  11. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  12. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
26 more
  1. 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
  2. 2017One pixel attack for fooling deep neural networkscited by 579
  3. 2014Understanding Deep Image Representations by Inverting Themcited by 399
  4. 2017Methods for Interpreting and Understanding Deep Neural Networkscited by 342
  5. 2015Understanding Neural Networks Through Deep Visualizationcited by 332
  6. 2018Learning Not to Learn: Training Deep Neural Networks with Biased Datacited by 255
  7. 2020Understanding the Role of Individual Units in a Deep Neural Networkcited by 254
  8. 2016Synthesizing the preferred inputs for neurons in neural networks via deep generator networkscited by 181
  9. 2016Generating Visual Explanationscited by 156
  10. 2020REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasetscited by 136
  11. 2018Turning a Blind Eye: Explicit Removal of Biases and Variation from Deep Neural Network Embeddingscited by 114
  12. 2016Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networkscited by 106
  13. 2015A Deeper Look at Dataset Biascited by 86
  14. 2021A Comprehensive Study on Face Recognition Biases Beyond Demographicscited by 86
  15. 2020Towards causal benchmarking of bias in face analysis algorithmscited by 64
  16. 2019Discrimination in the Age of Algorithmscited by 51
  17. 2017Interpreting CNN Knowledge via an Explanatory Graphcited by 49
  18. 2017Cognitive Psychology for Deep Neural Networks: A Shape Bias Case Studycited by 46
  19. 2020InsideBias: Measuring Bias in Deep Networks and Application to Face Gender Biometricscited by 31
  20. 2017Examining CNN Representations with respect to Dataset Biascited by 27
  21. 2019Deep Learning for Face Recognition: Pride or Prejudiced?cited by 25
  22. 2019Algorithmic Discrimination: Formulation and Exploration in Deep Learning-based Face Biometricscited by 19
  23. 2019SensitiveNets: Learning Agnostic Representations with Application to Face Imagescited by 17
  24. 2020Facial Expressions as a Vulnerability in Face Recognitioncited by 11
  25. 2021Towards Measuring Bias in Image Classificationcited by 6
  26. 2020Symbolic AI for XAI: Evaluating LFIT Inductive Programming for Fair and Explainable Automatic Recruitmentcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 2; 2023: 3; 2024: 2; 2025: 2; 2026: 1.

11 papers.

  1. 2023Human-Centric Multimodal Machine Learning: Recent Advances and Testbed on AI-based Recruitmentcited by 32
  2. 2021Anatomizing Bias in Facial Analysiscited by 18
  3. 2024Is My Data in Your AI? Membership Inference Test (MINT) applied to Face Biometricscited by 12
  4. 2025PBa-LLM: Privacy- and Bias-aware NLP using Named-Entity Recognition (NER)cited by 10
  5. 2022FaceQgen: Semi-Supervised Deep Learning for Face Image Quality Assessmentcited by 4
  6. 2022A survey of Identification and mitigation of Machine Learning algorithmic biases in Image Analysiscited by 3
  7. 2023Data augmentation and explainability for bias discovery and mitigation in deep learningcited by 2
  8. 2023A survey on bias in machine learning researchcited by 2
  9. 2024Comprehensive Equity Index (CEI): Definition and Application to Bias Evaluation in Biometricscited by 2
  10. 2025Balancing Tails when Comparing Distributions: Comprehensive Equity Index (CEI) with Application to Bias Evaluation in Operational Face Biometricscited by 2
  11. 2026Unraveling Machine Behavior by Multi-Level Bias Analysis and Detection: Methodology and Application to Computer Visioncited by 2