When Algorithms are Biased

Data Science is increasingly driving many of the choices available to us. Algorithms help online companies recommend content, products and advertisements. Retailers are using them to target people for offers. They narrow down our romantic options and filter and recommend job candidates for corporations. Data algorithms are expected to be more objective than humans, but unfortunately can have bias and even discrimination built into them. Sometimes these outcomes reflect the unconscious biases of engineers; other times such errors are baked into machine learning itself. This panel will discuss the examples of algorithms going wrong and the causes for bias in data science and machine learning. Panelists will explore solutions to this problem, including the need for diversity in the promising field of data science and statistics, as well as approaches companies can implement to weed it out.

Speakers:

Elizabeth Dwoskin, Wall Street Journal (moderator)

Andrew Moore, Dean of Carnegie Mellon’s School of Computer Science

Sophie Lebrecht, CEO at Neon-Labs

Adeyemi Ajao, VP Technology Strategy at Workday

Rachel Poulsen, Director of Data Science & Measurement at TiVo

Meet the panel:

Elizabeth Dwoskin, Wall Street Journal (moderator)

http://photos2.meetupstatic.com/photos/event/8/f/0/5/600_441516613.jpeg

Elizabeth Dwoskin writes about technology from The Wall Street Journal's San Francisco bureau. She covers the growing world of big data across many industries, from health care to politics to advertising. Topics include privacy, innovation, and how predictions and algorithms are changing the way we live. Previously, Ms. Dwoskin was a staff writer for Bloomberg Businessweek magazine, covering policy and politics in Washington, D.C.

Ms. Dwoskin has a B.A. from Columbia University and has a master’s degree from the Columbia University Graduate School of Journalism. She is a two-time finalist for the Livingston Award for Young Journalists.

Andrew Moore, Dean of Carnegie Mellon’s School of Computer Science

http://photos4.meetupstatic.com/photos/event/3/b/4/3/600_441615171.jpeg

Andrew W. Moore. PhD, a distinguished computer scientist with expertise in machine learning and robotics, became dean of the Carnegie Mellon University School of Computer Science in August 2014. He had previously served as a professor of computer science and robotics before taking a leave of absence to become founding director of Google's Pittsburgh engineering office in 2006.

Moore's research interests broadly encompass the field of "big data"--applying statistical methods and mathematical formulas to massive quantities of information, ranging from Web searches to astronomy to medical records, in order to identify patterns and extract meaning from that information. His past research has also included improving the ability of robots and other automated systems to sense the world around them and respond appropriately.

Sophie Lebrecht, CEO at Neon-Labs

http://photos1.meetupstatic.com/photos/event/8/f/a/a/600_441516778.jpeg

Sophie Lebrecht received her PhD in Cognitive Science at Brown University in 2011. She continued to pursue her work on the neural basis of visual preference for everyday objects at Carnegie Mellon University and CMU Silicon Valley, eventually transitioning her research into the foundation of Neon, where she is CEO and co-founder. At Neon, Sophie and her team bring science and automation to the problem of finding engaging images online for media, sports, and marketers.

Neon has been recognized for innovation, winning World Economic Forum Tech Pioneers award, The Edison Award for Innovation, funded by The National Science Foundation, and featured in The White House, Forbes, Dow Jones, and The Wall Street Journal. Sophie’s innovative work was recognized by Fast Company in their Most Creative People 2015 list.

Adeyemi Ajao, VP Technology Strategy at Workday

http://photos1.meetupstatic.com/photos/event/5/2/7/2/600_441621106.jpeg

Adeyemi Ajao was co-founder & CEO of Tuenti.com the "Facebook of Spain" (acquired by Telefonica in 2010 for $100 million) and most recently of Identified.com (acquired by Workday). He is also a co-founder of Cabify.com (on-demand transportation in Spain and LatAm) and Jobandtalent.com (online recruiting platform in Spain & UK).

He is currently Vice President of Product Technology Strategy at Workday, focusing on Workday's analytic products strategy as well its data science & data engineering efforts as well as running Workday Ventures.

He earned a J.D. Law and Masters Economics and Finance degree from ICADE in Madrid and an M.B.A. and certificate in machine learning from Stanford.

Rachel Poulsen, Director of Data Science & Measurement at TiVo

http://photos2.meetupstatic.com/photos/event/8/f/9/2/600_441516754.jpeg

Rachel studied Statistics, Communications, and Sociology. While she enjoyed the problem-solving aspect of Math and Statistics, she also wanted the social interactions Communications and Sociology provided.

She has worked at L-3 Communications, American Express and TiVo as a statistician digging into various data sources to find insights. She spent some time at Silicon Valley Data Science -- a start-up consulting company -- working on building data strategies for a variety of companies.

She is currently the Director of Data Science at TiVo where working to build a big data platform and strategy there.

Agenda:

6:00 - 6:45pm: Registration, Networking
6:45 - 7:00pm: Introductions
7:00 - 7:45pm: Presentation
7:45 - 8:00pm: Q&A session
8:00 - 8:30pm: Networking

PARKING
In order to park in the Visa Metro 1 Garage, it is crucial that you RSVP to this event. Please provide your first and last name to security at the garage, as they will have a VIP list. If your name is NOT on the list, reference the event. Alternative parking can be found across the street, however, be aware that you may be ticketed.