Group items tagged

The blind application of machine learning runs the risk of amplifying biases present in data. 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. We show that even word embeddings trained on Google News articles exhibit female/male gender stereotypes to a disturbing extent. This raises concerns because their widespread use, as we describe, often tends to amplify these biases. Geometrically, gender bias is first shown to be captured by a direction in the word embedding. Second, gender neutral words are shown to be linearly separable from gender definition words in the word embedding. Using these properties, we provide a methodology for modifying an embedding to remove gender stereotypes, such as the association between between the words receptionist and female, while maintaining desired associations such as between the words queen and female. We define metrics to quantify both direct and indirect gender biases in embeddings, and develop algorithms to "debias" the embedding. Using crowd-worker evaluation as well as standard benchmarks, we empirically demonstrate that our algorithms significantly reduce gender bias in embeddings while preserving the its useful properties such as the ability to cluster related concepts and to solve analogy tasks. The resulting embeddings can be used in applications without amplifying gender bias.

The Well Said project is a five-year initiative led by Yle and the Timeout Foundation. The purpose of the project is to strengthen the best aspects of Finnish conversational culture and create safe environments that foster discussion and conversations. At the same time, the project aims to increase mutual trust and understanding among Finns. The goal of the project is to raise the level of Finnish conversational culture and inspire people to participate in respectful discussions and conversations that take multiple perspectives into account. The Well Said project will run from 2021 to 2026 and include a wide range of partners. The project will be highlighted prominently in Yle's contents and on its partners' numerous platforms.