shiwaneeg

I am a marketing intern at Valuefirst Digital Media. I write blogs on AI, Machine Learning, Chatbots, Automation etc for House of Bots. ...

Full Bio 
Follow on

I am a marketing intern at Valuefirst Digital Media. I write blogs on AI, Machine Learning, Chatbots, Automation etc for House of Bots.

Why is there so much buzz around Predictive Analytics?
527 days ago

Changing Scenario of Automation over the years
528 days ago

Top 7 trending technologies in 2018
529 days ago

A Beginner's Manual to Data Science & Data Analytics
529 days ago

Artificial Intelligence: A big boon for recruitment?
530 days ago

Top 5 chatbot platforms in India
38865 views

Artificial Intelligence: Real-World Applications
24789 views

Levels of Big Data Maturity
15267 views

Challenges of building intelligent chat bots
14331 views

Chatbots' role in customer retention
13980 views

IBM's Deep Learning as a Service uses the cloud to democratize AI for developers

By shiwaneeg |Email | Mar 21, 2018 | 8553 Views

At its Think 2018 conference in Las Vegas on Tuesday, IBM rolled out its Deep Learning as a Service (DLaaS) program for artificial intelligence (AI) developers.

The service is available through Watson Studio, and is aimed at helping developers run hundreds of deep learning training models at the same time while building out their neural networks, according to a press release. The firm has been working on the service since at least the middle of last year, according to a white paper from IBM researchers.

By using the power of the cloud to deliver AI capabilities like deep learning, IBM and other vendors that have similar services are democratizing access to these tools. Since companies won't have to build and maintain costly hardware to experiment with deep learning, it could mean more companies are able to leverage the power of AI in their own products and services.

"DLaaS provides developers the flexibility to use popular deep learning libraries, such as Caffe, Torch, and TensorFlow in the cloud in a scalable and resilient manner with minimal effort," the white paper said. "The platform uses a distribution and orchestration layer that facilitates learning from a large amount of data in a reasonable amount of time across compute nodes."

The DLaaS product also features a resource provisioning layer, according to the paper. This allows for jobs to be managed across heterogeneous resources, e.g. using both graphics processing units and central processing units to run workloads.

Users simply collect and organize their data before uploading it to the service. IBM will automatically start the training and stop it when it is completed, provisioning resources so the user doesn't overpay. This automation could help save time in the deep learning process as well.

In addition to The DLaaS initiative, IBM also announced that new IBM Power Servers, designed for AI, will be heading toward certain cloud data centers, the release said. IBM Cloud Object Storage will now also include a high-speed data transfer option called IBM Aspera for moving data to the cloud.

Additionally, the release said, a new IBM Cloud Private Data offering will ingest data "100x faster than competitors and makes data ready for AI," and another new SQL service will make data analytics easier.

Security also got a boost at Think, as IBM unveiled that its Hyper Protect Family of services achieved FIPS 140-2 Level 4 certification, the release said. New capabilities for cloud apps, like IBM Cloud Security Advisor, will also help security, and a new Cloudflare partnership will also improve the safety of certain websites.

To enable GDPR compliance, New Relic is delivering services out of IBM Cloud's data center in Frankfurt, Germany. Also, the release noted, new "features for the IBM Cloud Private platform enable enterprises to create their own clouds in internal or external data centers." Deeper SAP integrations and an IBM Cloud Container Service on Bare Metal were also announced.


This article was originally published in TechRepublic

Source: TechRepublic