Brute force feature engineering - Yam Peleg, Deep Trading скачать в хорошем качестве

Brute force feature engineering - Yam Peleg, Deep Trading

This lecture was part of the AutoML conference, organized by the MDLI community. Link: https://bit.ly/AutoMLConf The importance of creating the proper features cannot be overstated because a machine learning model can only learn from the data we give to it. Extracting as much information as possible from the available datasets is crucial to creating an effective solution. However, manual feature engineering is a tedious task and is limited by both human imagination - there are only so many features we can think to create - and by time - creating new features is time-intensive. Automated feature engineering methods aim to help the data scientist with the problem of feature creation by automatically building hundreds or thousands of new features from a dataset. although it will not replace data scientists, it allows them to focus on more valuable parts of the machine learning pipeline, such as delivering robust models into production. In this talk, we will look at an exciting development in the field of AutoML: Automated feature engineering! We will review the different approaches in terms of performance as well as time-saving and also briefly explain how we utilize those approaches in real-time production systems.