News
AWS Connects Machine Learning Scientists with Customers
Amazon Web Services (AWS) customers looking to undertake a machine learning project for the first time can now tap AWS' in-house machine learning experts to help them through the process.
Earlier this month, the company announced the launch of the Amazon ML Solutions Lab, a new program that lets AWS customers receive face-to-face project guidance and training directly from AWS' own machine learning scientists.
"Amazon ML Solutions Lab engagements are designed to take you through the full process of implementing machine learning, combining educational workshops and boot camps, advisory professional services, and hands-on help building custom models with a your own data," said Vinayak Agarwal, senior product manager for AWS Deep Learning, in a blog post describing the program.
According to Agarwal, each engagement under the Amazon ML Solutions Lab has three components. First, AWS scientists will work with the customer to identify the areas in their business that can benefit from machine learning.
Then, to facilitate those opportunities, the scientists will help the customer develop prototypes of machine learning models customized around their business data. According to Agarwal, these models can be as varied as the AWS scientists' areas of expertise, which include "fulfillment and logistics, personalization and recommendation, fraud prevention, anomaly detection, transactional risk management, security, forecasting, capacity planning, text and mobile analytics and supply chain optimization."
The third component is education. AWS scientists will provide formal machine learning training in the form of bootcamps, workshops and tutorials.
The program stresses the importance of face-to-face interactions between customers and AWS scientists. Each customer in the program can choose to either work with AWS scientists in the company's Seattle headquarters, or to host AWS' scientists in their own offices. There are also weekly meetings to assess the progress of their machine learning project.
A typical engagement takes between three to six months, according to this FAQ. However, for customers that are a little bit more advanced on their machine learning knowledge, AWS provides an "Express" version of the lab that spans four or five weeks. This accelerated program consists of a one-week educational bootcamp at the AWS campus, followed by "intensive problem-solving and machine learning model building," as Agarwal described it.
The Amazon ML Solutions Lab is now generally available globally, though the Express program is only available within the United States. More information on the program is available here.