Automated machine learning is making data science and machine learning accessible to more people. An emerging area of automated machine learning is ensemble generation, a process where multiple algorithms are combined automatically that, together, provide better results than each individual algorithm on its own. Many of the machine learning contests, such as those on Kaggle, are won by ensemble algorithms manually assembled by teams of data scientists.

Most AutoML products currently available don’t attempt to build ensembles, instead assuming a user will prepare them, a process that takes a considerable amount of time and specialized knowledge. Since this level of knowledge is beyond many developers or business analysts, the use of this powerful tool has historically been limited to those with advanced knowledge of machine learning and data science.

In this webinar, we will examine what an ensemble is, and common ways they are created with specific references to how Auger handles the task.