Empowering Builders – unlocking the power of machine learning
This week, during the first ever dedicated machine learning keynote at AWS re:Invent, Swami Sivasubramanian, Vice President of Amazon Machine Learning, announced a number of features and services with the aim of making machine learning even more accessible for all developers.
Swami announced nine new capabilities for Amazon SageMaker, our fully managed machine learning service that removes challenges from each stage of the machine learning process, making it radically easier and faster for everyday developers and data scientists to build, train, and deploy machine learning models for virtually any use case. These services bring together powerful new capabilities like faster data preparation, a purpose-built repository for prepared data, workflow automation, greater transparency into training data to mitigate bias and explain predictions, distributed training capabilities to train large models up to two times faster, and model monitoring on edge devices.
Machine learning is becoming more mainstream, but it is still evolving at a rapid clip. With all the attention machine learning has received, it seems like it should be simple to create machine learning models, but it isn’t.
In order to create a model, developers need to start with the highly manual process of preparing the data. Then they need to visualize it in notebooks, pick the right algorithm, set up the framework, train the model, tune millions of possible parameters, deploy the model, and monitor its performance. This process needs to be continuously repeated to ensure that the model is performing as expected over time. In the past, this process put machine learning out of the reach of all but the most skilled developers. However, Amazon SageMaker has changed that.
Today, more than 100,000 AWS customers use machine learning, from personalised customer experiences to developing personalised pharmaceuticals. Domino’s uses machine learning for predictive ordering, and to meet their goal of 10 minutes or less for pizza delivery. Nike uses it to provide product recommendations to deliver a more relevant experience in wholesale, while Formula 1 incorporates machine learning in car design on 550 million data points.
Machine learning is already helping companies make better and faster decisions. And while we’re encouraged by the applications we’re seeing today, we know we have only just started to scratch the surface of what is possible. I, for one, am excited to see how these new services will be used by our customers in the UK and Ireland to have a greater positive impact on their businesses, and more widely on society.
For example, in education, Dublin-headquartered, Terminalfour, is using AWS machine learning to helps universities and colleges to deliver exams and assessments online, at scale and securely. Meanwhile BetterExaminations, part of the Terminalfour group, is helping educators to securely create, approve, manage, deliver, proctor, invigilate and mark examinations through a single cloud-based eAssessment platform which runs on AWS. The platform can detect patterns that indicates attempts to cheat and automatically mark multiple-choice and maths exams. During an unprecedented time, like the COVID-19 pandemic, this really matters.
If you want to hear more about this topic, take a few minutes to watch this interview between Ronald van Loon and Julien Simon, Principal Advocate, ML/AI at AWS. There’s much more to come from AWS re:Invent 2020, but don’t worry if you miss a session or a keynote. You can catch up at any time by logging on to the AWS re:Invent website and looking through the Session Catalogue.
Don’t forget to tune in to watch Senior Vice President of Global Infrastructure and Customer Support Peter DeSantis’ keynote on Thursday 10th December, 3:30-5:30pm GMT. Peter will explain how AWS has optimised its cloud infrastructure to run of the world’s most demanding workloads and give your business a competitive edge. I look forward to seeing you there.