Andrew Weeks - Building fairer models for finance | PyData Global 2020

Share
Embed
  • Published on:  Tuesday, January 5, 2021
  • Talks
    Fairness and bias are rightly hot topics in ML because of their impact on people’s lives. It’s particularly important in finance, where bad decisions can limit the ability to participate in society. We look at how fairness can be defined, and how the law defines it. We explore how to detect bias, the trade off between bias and performance, and how to build fairer models that still perform well.

    Speaker
    I’m a product-focused data scientist at Aire, a fintech startup and credit reference agency. I’ve been part of the startup scene in London for nearly a decade.

    At Aire I’ve been responsible for shaping products for the UK and US markets from an early stage, building credit risk models and other insights, and worked closely on our governance framework and fairness processes.

    I try to spend as much time as I can on understanding the problem. It’s usually a lot messier and more complex than I realise, but it’s key to building solutions that have a real impact!

    www.pydata.org

    PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.

    PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases. 00:00 Welcome!
    00:10 Help us add time stamps or captions to this video! See the description for details.

    Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/numfocus/YouTubeVideoTimestamps
Loading...

Comment