"How to Choose a Machine Learning Algorithm: A Machine Learning Approach"
Today's data scientists are facing two major challenges: an explosive growth in data volume, and an equally explosive growth in machine learning algorithms for analyzing that data. By the time you finish reading this sentence, twelve new algorithms for classification, clustering, and time series mining will have been published in the proceedings of a prestigious machine learning conference. The FOMO that this situation has engendered has become debilitating for many data scientists, who find themselves spending so much time reading papers that they no longer have time for their day jobs (i.e., counting things and speaking at conferences.) We're going to take an hour and talk about how to free yourself from this Chinese Restaurant process: how to choose an algorithm that is a good fit for your problem, the relationship between systems and algorithms, the bane (and job security) of tuning parameters, and how to know if it's time to go hire the grad student who wrote that one paper you really liked.
Join us for another Uber Tech Talk! Doors will open at 6PM, food and drinks will be provided. We look forward to seeing you there!
6PM: Doors open
6-7PM: chat, drink, mingle
7-8PM: Talk, Q&A
8PM-??: more chatting, drinking and mingling
Join millions of people on Eventbrite.