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.
View on arXiv