Page 38 - MSDN Magazine, November 2019

Saying Goodbye
The ethical and political issues around facial recognition may be complicated, but the algorithms enabling basic face detection and recognition are not. In fact, much of what this article explored could’ve been easily accomplished with the Cognitive Services Face API via a few simple REST calls. However, I feel it’s important for software engineers to have a deeper knowledge of the algorithmic underpinnings of these tools and to appreciate the work that went into creating the Face API. Additionally, knowing more about the underlying mathematical principles can help developers identify edge cases they may encounter.
In this article I touched upon the rich and fascinating subject of facial recognition. This is very much at the forefront of AI research. Much is still being done in this field to reduce the rate of false positives and mitigate biases. And researchers continue to make discoveries, for instance the recent finding that certain patterns and colors in face paint can hinder many face detection systems (bit.ly/2mkn0hW). n
Frank La Vigne works at Microsoft as an AI Technology Solutions Professional where he helps companies achieve more by getting the most out of their data with analytics and AI. He also co-hosts the DataDriven podcast. He blogs regularly at FranksWorld.com and you can watch him on his YouTube channel, “Frank’s World TV” (FranksWorld.TV).
Thanks to the following Microsoft technical expert for reviewing this article: Andy Leonard
These last few years have been an interesting personal and professional journey for me, as I transitioned from a smart client developer to a machine learning engineer. I discovered that much of the training material in the artificial intelligence (AI) space was geared toward academic researchers and those with extensive experience in advanced mathematics. There just wasn’t much for software engineers to grab hold of. This inspired me to retire the Modern Apps column and its coverage of UWP development, and re-launch as Artificially Intelligent. The goal: To explore the realms of data science and AI from a software engineer’s point of view.
This column will be coming to an end, but I plan to continue my work making AI and data science more approachable to soft- ware developers. I’ll continue to post articles on my blog and to podcast at DataDriven. Recently, I started hosting virtual summits, one-day virtual events focused on a particular topic. And I have other projects in the works to continue my mission of developer training and empowerment. For a full list of activities and resources, visit franksworld.com/msdn for offers, links and discounts just for readers of this column.
It has been a distinct honor and privilege to write for MSDN Magazine these last several years. I’ve been amazed and delighted to be recognized for my column by people at an event or customer meeting, and to hear how much they learned from the column over the years. Finally, I would like to extend my deep gratitude to Rachel Appel, whose column I took over in 2016, and to Michael Desmond for being such a great (and patient) editor.