Exploring the Linear Subspace Hypothesis in Gender Bias Mitigation

Add to your list(s) Download to your calendar using vCal

• Francisco Vargas Palomo
• Monday 07 December 2020, 16:30-17:00
• virtual ZOOM meeting ID: 263 591 6003, Passcode: 000042, https://us02web.zoom.us/j/2635916003?pwd=ZlBEQnRENGwxNmJGMENGMWxjak5nUT09.

If you have a question about this talk, please contact Bingqing Cheng .

Paper: https://www.aclweb.org/anthology/2020.emnlp-main.232.pdf

Abstract: Bolukbasi et al. (2016) presents one of the first gender bias mitigation techniques for word embeddings. Their method takes pre-trained word embeddings as input and attempts to isolate a linear subspace that captures most of the gender bias in the embeddings. As judged by an analogical evaluation task, their method virtually eliminates gender bias in the embeddings. However, an implicit and untested assumption of their method is that the bias subspace is actually linear. In this work, we generalize their method to a kernelized, non-linear version. We take inspiration from kernel principal component analysis and derive a non-linear bias isolation technique. We discuss and overcome some of the practical drawbacks of our method for non-linear gender bias mitigation in word embeddings and analyze empirically whether the bias subspace is actually linear. Our analysis shows that gender bias is in fact well captured by a linear subspace, justifying the assumption of Bolukbasi et al. (2016).

This talk is part of the Machine learning in Physics, Chemistry and Materials discussion group (MLDG) series.

This talk is included in these lists:

• Hanchen DaDaDash
• Lennard-Jones Centre external
• Machine learning in Physics, Chemistry and Materials discussion group (MLDG)
• virtual ZOOM meeting ID: 263 591 6003, Passcode: 000042, https://us02web.zoom.us/j/2635916003?pwd=ZlBEQnRENGwxNmJGMENGMWxjak5nUT09

Note that ex-directory lists are not shown.