IBM Corp. today introduced new versions of its Cloud Pak for Data and Cloud Pak for Automation products that will enable enterprises to harness deep learning models in their operations more easily.

The Cloud Pak product family is a set of software solutions designed to streamline a variety of tasks ranging from cybersecurity to data analytics. The entire lineup is based on Red Hat OpenShift. Thanks to OpenShift, the solutions can run both in the public cloud and on on-premises infrastructure. Deep learning is a branch of artificial intelligence that uses artificial neural networks to learn from massive amounts of data.

IBM Chief Executive Arvind Krishna (pictured) in October named the hybrid cloud and artificial intelligence as core pillars of the company’s revenue growth strategy. The Cloud Pak product family sits at the intersection of the two trends. As a result, it may take on a more important role in IBM’s roadmap going forward as the company works to grow its software business into a bigger slice of total revenues. 

Cloud Pak for Data, the first product for which IBM announced updates today, is designed to help companies organize business records and analyze them. Data scientists can now build deep learning models to help them analyze information using a tool called Watson Machine Learning Accelerator that has been added to the product. The tool provides an “end-to-end” set of features for model training and related tasks such as optimizing neural networks’ hardware efficiency. 

The update also brings a couple of more specialized AI features. One is support for federated learning, a privacy technique that makes it possible to train an AI model on a dataset without having to move that dataset to another location. Additionally, IBM is adding features to simplify the creation of AI training datasets that contain information from multiple sources.

“The latest version of IBM Cloud Pak for Data automates and helps many of the time-consuming administration and maintenance tasks of delivering data and AI at scale,” stated Hemanth Manda, executive director of IBM’s Data and AI unit.

The other product receiving an upgrade is Cloud Pak for Automation. It’s a collection of tools that companies can use to automate back-office workflows such as moving records between disparate business applications.

The tent pole enhancement is that the product now incorporates the robotic process automation technology IBM obtained through its recent acquisition of WDG Automation Inc. Ltd. WDG’s technology can perform repetitive back-office chores by capturing the clicks and other actions that a worker makes while performing a task in an application, then replicating their workflow.

In the same spirit, IBM has added new deep learning features for automatically extracting data from documents. The update comes just days after Google LLC introduced a new AI offering designed to pull data from paperwork such as forms. Document processing has emerged a focus for other major industry players as well, including cloud operators and RPA providers, because it’s a use case that’s relevant to practically every enterprise, which means there’s a big addressable market.

Photo: SiliconANGLE