AutoML (automated machine learning) has become a hot topic over the past few years. Abid Ali Awan joins the show to share his approach to AutoML, when and how to utilize it compared to classic approaches. Ben and Abid also discuss open-source vs. proprietary platforms.
What is AutoML?
Automated machine learning provides methods and processes to make machine learning available for non-machine learning experts, to improve efficiency of machine learning and to accelerate research on machine learning.
2 levels of implementation:
Blackbox AutoML can do one, or all of the things for feature selection with a statistical outset and self optimizing outcome.
Whitebox AutoML exposes the code to explain how it behaves and allows you to produce predictions as to the influencing variables.
Leveraging open-source toolkits vs. proprietary
“The output is the model, the input is the data, and you can use that model to predict anything according to your business problem.” - Abid Ali Awan