Latest AI product news.
Machine Learning in the Cloud
Use the cloud to develop and deploy rapidly and at scale. Thoughts on AWS, Google, Microsoft Azure and others.
AI for Product ProfesSionals
Best practices, latest trends & developments, top use cases and essentials for managing AI products.
Get up to speed on the basics – what machine learning is, how it works and getting started.
Product management & machine learning
AI Product Management
– Evolution or Revolution?
In this article, I argue that as a Product Manager you’ll need one of three levels of expertise: “AI Aware”, “AI Enabler”, “AI Expert”, in order to provide the leadership needed to evolve your product and services as AI becomes ubiquitous.
A ‘second wave’ of AI will extend to reach almost every sector and industry. Maybe machine learning is already in your product – but do you understand all of the implications?
The ‘AI product chasm’
Even the software industry, although leading the way, is lacking Product Managers with the right skills to fully address the AI opportunity.
today’s Product Roles
Do you know your Product Manager from your Product Owner? There’s a framework that can help…
accessing machine learning
Build it yourself, partner, use cognitive APIs or increasingly powerful AutoML.
A Step by step guide
Building a Simple AI Photo App on AWS
We’re going to use some clever AWS features to build an AI powered image analytics app, providing machine learning derived insights on photos in near real time. What’s more, we’re going to use ‘serverless’ components, meaning you should be able to do get this up and running on your images in a couple of hours or less!
About the APP
Load photos into your S3 bucket and view AI image analytics in QuickSight.
An S3 alert triggers an AWS Lambda function for each image as it’s uploaded, processing them in parallel.
Amazon’s image AI service returns object descriptions, locations and text content for each image.
Amazon QuickSight accesses the JSON files created by Lambda from Rekognition’s responses, via Athena, and provides an analytics dashboard.
Automated machine learning
An Introduction to AutoML
In this article, I use Kaggle to help illustrate components of a machine learning workflow. I then look at three very different AutoML providers & solutions, contrasting them to help clarify what AutoML is. Then I give them a go…
AWS SaGeMaker Autopilot
How SageMaker Autopilot fits with other Amazon AI services, and a quick test drive. AutoML for developers.
A look at Google’s AutoML, before running Google AutoML Tables on a Kaggle dataset. AutoML for the non technical.
H2O.ai Driverless AI
A look at H2O.ai’s very powerful and intuitive AutoML tool. AutoML for Data Scientists.
I define two distinct AutoML types: ‘Closed’ and ‘Open’ and map them to the end to end machine learning process.
Understanding Federated Learning Types
Navigate the strange world of Federated Learning, where you can build machine learning models using data you don’t own and can’t see.
Federated Learning is used today – by Google to improve Android user experience and by banks in China to improve credit scoring and anti-money laundering detection. Models are built with combined knowledge, but no data sharing.
Model or Data Centric?
Current research papers and case studies use a range of terms to describe different Federated Learning contexts. See how these terms fit together.
Horizontal vs Vertical
How data is split affects how and where Federated Learning will be implemented.
This article was written with help from the OpenMined community and is published on their blog. OpenMined are extending Facebook AI’s PyTorch with PySift and PyGrid, helping make practical privacy preserving AI a reality.
Building on June’s AWS Machine Learning Questions, this month we’re adding free & original questions to help you with Amazon’s Data Analytics – Specialty certification.