Over the past few years, we\u2019ve seen that artificial intelligence (AI) and machine learning (ML) provide us with new opportunities, but they also raise new challenges. Most notably, these challenges have highlighted the various ways in which AI systems can promote unfairness or reinforce existing societal stereotypes. While we can often spot fairness-related harms in AI systems when we see them, there\u2019s no one-size-fits-all definition of fairness that applies to all AI systems in all contexts. Additionally, there are many reasons why AI systems can behave unfairly. In this session, we discuss strategies for mitigating fairness-related harms and the research questions that are raised when working on fairness in AI systems. We cover fairness in recommendation systems, present how checklists can support fairness in the AI lifecycle, and discuss research questions on the challenges of measuring computational harms and the trade-offs in choosing an appropriate fairness metric.<\/p>\n
Introduction<\/strong> Fairness via post-processing in web-scale recommender systems<\/strong> Designing checklists to support fairness in the AI lifecycle<\/strong> Challenges to the discovery and measurement of computational harms<\/strong>
\nSpeaker: Amit Sharma, Senior Researcher, Microsoft Research India<\/p>\n
\nSpeaker: Kinjal Basu, Tech Lead for Responsible AI, LinkedIn<\/p>\n
\nSpeaker: Michael Madaio, Postdoctoral Researcher, Microsoft Research NYC<\/p>\n
\nSpeaker: Alexandra Olteanu, Principal Researcher, Microsoft Research Montr\u00e9al<\/p>\n