On February 2, the Microsoft New England Research & Development (NERD) Center hosted and sponsored the “Hacking Bias & Discrimination Ideation Session,” a precursor to the Harvard-MIT chapter of the Women in Data Science Conference.
This six-hour event was designed to tackle the presence of bias in data science models and algorithms through discussion and brainstorming of the following topics:
- Gender Bias in Word Embeddings: http://www.wordbias.org/
- Identifying Gender Bias in Performance Reviews
- Neuroinformatics Research Group Understanding Response Bias
- Bias in ML algorithms applied to healthcare
- Bias in Law Enforcement Predictions
- Consumer Protections in a Digital Age
- Hiring and Selection Models
- Who Has Political Power and How Do You Measure It?
If you’d like to attend an upcoming Hacking Bias and Discrimination event, please consider:
- Feb 25, 2017: MIT Sloan Breaking the Mold Hackathon
- April 28-29, 2017: MIT Alumni Hacking Discrimination Hackathon
- May 11, 2017: New England Machine Learning Day pre-event (at NERD)
The following day, more than 200 participants attended the Women in Data Science Conference held at NERD. Cathy Chute, Executive Director of the Harvard Institute for Applied Computational Science, and Elizabeth Sikorovsky, Executive Director of the MIT Institute for Data, Systems, and Society, delivered the opening remarks, followed by Elizabeth Bruce, Microsoft’s University Relations Lead, who presented a summary of the pre-event outcomes. Our very own Jennifer Chayes, Managing Director of Microsoft Research NE & NYC, delivered the keynote presentation, followed by talks from Microsoft Researcher Jennifer Listgarten, MIT professors, and data science industry professionals. Click Here for the conference agenda, abstracts and slides.
A recap video, via Zac Carpman: