Abstract

Abstract:

Given the impressive performance boosts provided by pre-trained models, the fields of computer vision and natural language processing are investing more heavily in social prediction tasks. However, such an dedication comes at the risk of several forms of ethical risks and harms to marginalized communities. In particular, machine learning for social and human data runs the risk of mirroring nineteenth century phrenologist methods. In this light, this talk discusses the question: what does it mean to predict morality on the basis of modern day phrenologist methods?

Bio:

Zeerak Talat is a post doctoral fellow at the Digital Democracies Institute at Simon Fraser University. Talat received a Ph.D. from the University of Sheffield. In the Ph.D., Talat worked on automated content moderation and how the practice of automating content moderation using machine learning revealed underlying the political economy of machine learning, displaying issues of access, equality, and ethical practices. Talat also founded and runs the Workshop on Online Abuse and Harms, which focuses on the technical and social developments of automated content moderation infrastructure. Talat is currently working on critical machine learning and the philosophy of machine learning, aiming to identify the specific underlying causes for why and how machine learning is a currently a marginalizing technology.

Topic: NLIP Seminar Time: Jun 13, 2022 12:00 PM London

Join Zoom Meeting https://cl-cam-ac-uk.zoom.us/j/92076366810?pwd=ZU1IV21IbnFLWldqcDU5Vi9wNUd5QT09

Meeting ID: 920 7636 6810 Passcode: 933141