Amazon Web Services Boosts Its Cloud Computing With Machine Learning

July 23, 2018 - 3 minutes read

machine learning app developerAt Amazon’s AWS (Amazon Web Services) Summit in New York City this past week, machine learning was a big focus for the Seattle-based tech giant, especially in regard to recently released cloud features for commerce and payments.

The Rise of ML in Every Industry

Artificial intelligence (AI) and machine learning (ML) development are booming, especially in e-commerce and payment, two of Amazon’s most important back-end features. The company hopes these new features will work without any built-in bias and without a need for human intuition or intervention.

ML wasn’t always the talk of the town; AI wins that title. But many people don’t know that ML is a subset of AI. And if you’ve heard of deep learning, that’s a specialized subset of ML (and therefore AI). Algorithms are built by humans who try not to let bias seep into the multitude of if-else statements in the code.

ML is growing bigger and gaining more traction as an important part of today’s technology — Bessemer Venture Partners will soon roll out a hefty $10 million early-stage seed program to invest in new ML startups that want to improve healthcare.

Matt Wood is the general manager for ML in Amazon’s AWS department. “Machine learning is in virtually every industry, and every company,” says Wood. He adds that Amazon alone saw a 250% uptick in developers using AWS to create ML algorithms. This revelation comes on the heels of last year’s release of new ML features that AWS customers can use in coding and training customized AI algorithms.

Saving Time, Energy, and Effort

At the Summit conference, representatives from AWS said the department is continuing to improve AI and ML features and add new ones. One such improvement aims to boost the speed of AWS SageMaker, a tool that helps customers test and subsequently deploy their AI and ML algorithms.

Wood says this will increase how much training data developers can feed the algorithm. In turn, he says, this has decreased time barriers to start training algorithms by up to 90%. By saving developers time, companies can save money and focus resources on building better algorithms.

AWS has also been training developers on ML basics so that they can introduce ML into their enterprise applications without having to spend time continuing education. As one of the first enterprise-level cloud-computing platforms, Amazon’s taking the lion’s share of enterprise clients compared to Microsoft and Google.

We hope this eventually improves customer experience on Amazon’s main shopping site as well. What are your thoughts on your favorite apps and websites rolling out ML into their prediction algorithms?

Tags: , , , , , , , , , , , , , , , , , ,