Pondering Predictability
Details
Event space and snacks sponsored by Moody's
Agenda:
6:30pm – 7:00pm Mingling and snacks
7:00pm – 7:10pm Artificial Intelligence at Moody’s Analytics - Developing next generation products leveraging advances in AI and machine learning by Cristina Pieretti
7:10pm – 8:00pm Pondering Predictability by Claudia Perlich
8:00pm – 8:30pm Networking
'Predictability' has dominated my professional life as a data scientist ever since I tried to build my first predictive model in Matlab back in 1995. And to this day, the question, why I sometimes succeed at building a model that predicts well, sometimes too well, and sometime fails continues to haunt me. There are many fascinating nuances to this: how the inherent predictability affects which algorithm performs better on a given problem, how to even measure something like 'predictability', how predictability can possibly create unethical biases, or how it can be a symptom and diagnosis for all kinds of mistakes in the data pipeline.
Claudia (http://dstillery.com/leaders/claudia-perlich/) leads the machine learning efforts that power Dstillery’s digital intelligence for marketers and media companies. With more than 50 published scientific articles, she is a widely acclaimed expert on big data and machine learning applications, and an active speaker at data science and marketing conferences around the world.