Mr. Jeff Barr begins to explain what exactly machine learning is, its concept and its impact on the business.
At 0.008 he states that at the AWS (Amazon Web Service) a lot of time has been reserved for research and investigating in this field of study for nearly 20 years. He adds on that a qualitative number of 1000 engineers work at the AWS in the field of machine learning trying to research and build the machine learning services.
At 0:16 he puts forward the statement that these machine learning services would allow the users to access the exact capabilities, scale and costs irrespective of the size of the organization.
At 0:30 Jeff begins to explain about the machine learning stack that comprises of three significant layers. He begins with the framework & interfaces layer which is most suitable for the experts to build their own machine learning models and to train them. Deep learning AMI would play a huge role in getting started with the development of machine learning models.
At 0:43 he puts forward the example of Zillow that trains nearly seven and a half million models for the computation of the correct evaluation. At 0:54 he states the disadvantage of this layer is that it is not just enough for building superior models.
At 1:01, he begins to talk about the platforms layer that is present above the frameworks layer. He adds on that the Amazon Sagemaker makes it very easy for developers to build, train and deploy models without specialized expertise.
At 1:16, he states the example of AWS customer Intuit, that uses the Sagemaker. They are responsible to perform the fraud detection, identify the theft and build models to make excellent financial decisions. At 1:30 he begins to explain the topmost layer namely the applications services layer. He adds on that the chief objective of this layer is to help the developers make calls to the API (Application Programming Interface) so that the machine learning services get added to the applications without training of their own models.
At 1:38 he states that in applications services layer, APIs are called to perform the image processing, voice processing, and video recognition and speech synthesis. At 1:50 he emphasizes that before beginning to build on the AWS, one must take advantage of the storage, security, database and messaging features to augment their machine learning applications. At 2:00 he states that the customers like Netflix, Pinterest, NFL build their applications in a similar manner to access all the benefits from the AWS.
At 2:09, he briefs more about the NFL. He explains that the NFS uses the AWS machine learning for analyzing the defensive routes as well as the player coverage across the multiple games and multiple seasons. At 2:18 he summarizes that machine learning is a transformative technology that can be used to improve the results for the customers and business. He finally encourages people to start learning the nuances of machine learning and build it on the AWS to reap its benefits.
- Are Your AWS Workloads Well-Architected? - March 2, 2019
- How AWS Serverless Has Streamlined a Car Factory’s Supply Chain - February 26, 2019
- Smart Apartments with Alexa & AWS IoT - February 24, 2019
- Using Machine Learning on AWS – Raising the Barr - February 22, 2019
- Mastering Chaos – A Netflix Guide to Microservices - February 19, 2019
- Microservices at Netflix – Lessons Learned - February 17, 2019
- Segment: Managing and Monitoring Over 16,000 Containers on ECS - February 13, 2019
- Astadia: Migrating Mainframe Workloads to AWS - February 12, 2019
- Nordea takes Open Banking beyond PSD2 - February 12, 2019