Measuring Model Biases in the Absence of Ground Truth
| Autor: | Ken Burke, Alex Bäuerle, Osman Aka, Christina Greer, Margaret Mitchell |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Contextual image classification business.industry Computer science Association (object-oriented programming) Computer Vision and Pattern Recognition (cs.CV) Rank (computer programming) Computer Science - Computer Vision and Pattern Recognition Pointwise mutual information computer.software_genre Machine Learning (cs.LG) Information extraction Bag-of-words model Identity (object-oriented programming) Artificial intelligence business Set (psychology) computer Natural language processing |
| Zdroj: | AIES |
| Popis: | The measurement of bias in machine learning often focuses on model performance across identity subgroups (such as man and woman) with respect to groundtruth labels. However, these methods do not directly measure the associations that a model may have learned, for example between labels and identity subgroups. Further, measuring a model's bias requires a fully annotated evaluation dataset which may not be easily available in practice. We present an elegant mathematical solution that tackles both issues simultaneously, using image classification as a working example. By treating a classification model's predictions for a given image as a set of labels analogous to a "bag of words", we rank the biases that a model has learned with respect to different identity labels. We use man, woman as a concrete example of an identity label set (although this set need not be binary), and present rankings for the labels that are most biased towards one identity or the other. We demonstrate how the statistical properties of different association metrics can lead to different rankings of the most "gender biased" labels, and conclude that normalized pointwise mutual information (nPMI) is most useful in practice. Finally, we announce an open-sourced nPMI visualization tool using TensorBoard. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|