Machine Learning Logistics

How do you get a machine learning system to deliver value from big data? Turns out that 90% of the effort required for success in machine learning is not the algorithm or the model or the learning - it's the logistics. Ted Dunning and Ellen Friedman identify what matters in machine learning logistics, what challenges arise, especially in a production setting, and they introduce an innovative solution: the rendezvous architecture. This new design for model management is based on a streaming approach in a microservices style. Rendezvous addresses the need to preserve and share raw data, to do effective model-to-model comparisons and to have new models on standby, ready for a hot hand-off when a production model needs to be replaced.