iocas.blogg.se

Using these properties, we provide a methodology for modifying an embedding to remove gender stereotypes, such as the association between the words receptionist and female, while maintaining desired associations such as between the words queen and female. Second, gender neutral words are shown to be linearly separable from gender definition words in the word embedding. Geometrically, gender bias is first shown to be captured by a direction in the word embedding. This raises concerns because their widespread use, as we describe, often tends to amplify these biases. We show that even word embeddings trained on Google News articles exhibit female/male gender stereotypes to a disturbing extent. Such a danger is facing us with word embedding, a popular framework to represent text data as vectors which has been used in many machine learning and natural language processing tasks. The blind application of machine learning runs the risk of amplifying biases present in data. Bibtex Metadata Paper Reviews Supplemental