Databricks Integrates Machine Learning and Data Teams with Launch of Databricks Machine Learning

With sophisticated AutoML capabilities and a new ML Feature Store, the collaborative platform provides a simple method for businesses to standardize the whole data and machine learning lifecycle at any scale.

FREMONT, CA: Databricks launched Databricks Machine Learning, a new data-native platform built on top of an open lakehouse architecture, today at the Data + AI Summit, the latest iteration of its industry-leading machine learning (ML) solution. With Databricks Machine Learning, the Databricks Lakehouse Platform's new and existing ML capabilities, combined into a collaborative, purpose-built experience that gives ML engineers all they need to build, train, deploy, and manage ML models from experimenting to production, peculiarly integrating information and the entire ML lifecycle.

Two new features, added to Databricks Machine Learning: Databricks AutoML is a machine learning tool that automates all of the tiresome tasks that data scientists currently have to undertake while allowing for sufficient control and transparency. In a system embedded in the enterprise's data engineering platform, Databricks Feature Store improves model feature discoverability, reuse, and governance.

Databricks Machine Learning gives the essential tools to each member of the data team in a single collaborative environment. Users can transition between Data Science / Engineering, SQL Analytics, and the new Machine Learning experiences to access tools and capabilities relevant to their typical workflow. Experiments, the Feature Store, and the Model Registry are all accessible from a new ML-focused start page in Databricks Machine Learning. Databricks Machine Learning, built on an open lakehouse foundation, allows customers to work with any data, at any scale, for machine learning, from traditionally structured tables to unstructured data such as videos and images. Also, to stream real-time applications and IoT sensors and move quickly through the ML workflow to get more models to production faster.

"Humana's machine learning platform, FlorenceAI, is enabling us to automate and accelerate the delivery lifecycle of ML solutions at scale. Databricks has been an essential underlying technology, with hundreds of our data scientists using the platform to deliver dozens of models in production, so that our teams are able to operate at orders of magnitude faster than before," said Slawek Kierner, Senior Vice President of Enterprise Data and Analytics at Humana.

All AutoML experiments are linked to the rest of the Databricks Lakehouse Platform, including MLflow, to keep track of all the pertinent parameters, metrics, artifacts, and models associated with each trial run, making it simple to compare models and deploy them to production.

The Databricks Feature Stores is the first of its type, developed in collaboration with a data and MLOps platform. The Feature Store's close interaction with the popular open-source frameworks Delta Lake and MLflow ensures that data saved in the Feature Store is open. Models trained using any machine learning framework can benefit from the Feature Store's interaction with the MLflow model format.