Machine Learning Done Wrong

Cheng-tao Chu ( @chengtaochu ) joins us this week to discuss his perspective on common mistakes and pitfalls that are made when doing machine learning. This episode is filled with sage advice for beginners and intermediate users of machine learning, and possibly some good reminders for experts as well. Our discussion parallels his recent blog post Machine Learning Done Wrong .

Guest

Cheng-tao Chu: I am most recently a Software Architect at Cubist (Point72) Investment ($91.3B AUM) where I lead the engineering initiatives to build the next generation of research/simulation/trading infrastructure for our intraday future trading strategies. Prior to Cubist, I headed various data infrastructure and machine learning initiatives at different companies including Tudor investment ($38.4B AUM), my own startup, Codecademy, Square, LinkedIn and Google where I had gain extremely in-depth yet diversified experience. Both LinkedIn and Square have since IPO'ed . My 2006 paper published in the prestigious NIPS proceedings, "Map-Reduce for Machine Learning on Multicore", has received over 1,600 citations and is the basis of the open source Apache Mahout library.