Google's Cloud AutoML makes it easier to build custom machine learning models

Google Cloud has introduced a new product called Cloud AutoML that helps businesses and developers with limited machine learning expertise easily build powerful custom machine learning systems.
"Currently, there are only thousands of deep learning researchers and a million data scientists across the world who can build advanced machine learning models. With Cloud AutoML, we want to help less-skilled developers build these models" said Fei-Fei Li, Chief Scientist of Google Cloud AI and ML.
With Cloud AutoML, businesses will be able to create a simple machine learning model in minutes to pilot their AI-enabled application or build out a full production-ready model within a day, said Jia Li, head of R&D, Google Cloud AI.
Among the first Cloud AutoML products being developed is Cloud AutoML Vision, a service that makes it easier to create custom machine learning models for image recognition.
The service provides a drag-and-drop interface that lets developers easily upload images, train and manage models, and then deploy those trained models directly on Google Cloud.
Li said the service was developed in collaboration with Google Brain and other Google AI teams and has been piloted with select clients like Disney and Zoological Society of London.
“Cloud AutoML’s technology is helping us build vision models to annotate our products with Disney characters, product categories, and colors. These annotations are being integrated into our search engine to improve the impact on guest experience through more relevant search results, expedited discovery, and product recommendations on shopDisney” said Mike White, CTO, and SVP, for Disney Consumer Products and Interactive Media.
The Internet giant plans to expand it to other forms like speech, natural language, and video over the year. "We will gradually roll it out to all of Google Cloud APIs by the end of this year" said Rajen Sheth, senior director of product management, Google Cloud AI.

Trending News

Why Amazon’s Flipkart bid is just not high enough

Selling without MRP, expiry dates to cost etailers

TCS hits $100Bn market cap: A new phase for India Inc?

Snapdeal posts Rs 4,647 crore net loss in 2016-17

Flipkart-Walmart deal: What it means for the etailer and the whole industry

NCLAT stays CCI penalty of Rs 136cr on Google in search bias case

Walmart may rope in Google parent Alphabet for the Flipkart deal

Internet shutdown cost Indian economy $3 billion from 2012-17: Study

Flipkart and Amazon plan mega summer sales in May

International sales now account for 48% of Twitter's revenues

70% of mobile phones buyers in India opt for non-cash payments

Amazon India expects groceries & household products to account for 50% of business in 5 yrs

India to come out with a E-commerce policy framework in 6 months

Online learning platform Career Anna raises Rs 3 Cr funding

Wal-Mart will offer to buy up to 86% of Flipkart

Agritech startups Sabziwala and LivLush merge their business under new entity Kamatan

Avail Finance lands $17.2M from Matrix Partners & Ola, Freecharge and Flipkart founders

RBI suggests tax sops, self-regulation to build fintech space

Swiggy hires new CEO for its Access Service, gets new CFO

Logistics company Delhivery registers 44% increase in FY17 revenues

WeWork to acquire one of the oldest social networks, Meetup

Qualcomm rejects Broadcom's $103 billion offer

EasyRewardz gets $2 million Series-A funding

'Anemic' iPhone 8 demand drags Apple shares lower

Lending platform Lenden Club gets Rs 3.5cr in Equity Investment

On festive sales, Flipkart says 65% clients from Tier-II cities