What is Automated Machine Learning (AutoML)?

Machine learning is a field of artificial intelligence that uses statistical techniques to give computer systems the ability to "learn" (e.g., progressively improve performance on a specific task) from data, without being explicitly programmed. This trained machine learning model can later be used to predict the probability of a future event within an acceptable reliability.

For example, suppose that you want to predict the customer churn for a customer. You can use your historic customer data with customers you kept and lost to train a model using machine learning techniques. Later, you can feed a new customer info (or a set of new customer data) to the model and predict their probability to be lost.

This is a two step process. First you train a model and then you use this trained model to predict (or score in machine learning terms) new data.

Machine Learning training and scoring

Although it looks straightforward, traditional machine learning process requires some intermediary steps.  A typical machine learning application, practitioners must apply the appropriate data pre-processing, feature engineering, feature extraction, and feature selection methods that make the dataset amenable for machine learning. Following those preprocessing steps, practitioners must then perform algorithm selection and hyperparameter optimization to maximize the predictive performance of their final machine learning model.

Traditional machine learning process.

Many of these steps are often beyond the abilities of non-experts and automated machine learning (AutoML) addresses this problem. It is the process of automating the end-to-end process of applying machine learning to real-world problems.

As Janakiram MSV writes in Forbes (Why AutoML Is Set To Become The Future Of Artificial Intelligence):

“AutoML focuses on two aspects – Data acquisition and prediction. All the steps that take place in between these two phases will be abstracted by the AutoML platform. Essentially, users bring their own dataset, identify the labels, and push a button to generate a thoroughly trained and optimized model that's ready to predict.”

Automated Machine Learning

For example, take a look at the predictive fraud analytics for insurance sector example below. The AutoML application used here is cloud based Enhencer Predictive Story Teller. After integrating claims data to the predictive system, predictive model creation is done with a few clicks without expertise in data science and statistics (you can watch a more detailed introduction and demo here : Predictive Analytics and Machine Learning Introduction - Customer Propensity Example)