I'm happy to say that I have a speaking engagement Friday October the 9th. It's a special part of the annual ACM SIGDOC 2020 conference. I'm part of a small group of speakers featured in their so-called Ignite session. Ignite talks are meant to be quick, energetic, and insightful briefs on an interesting subject.
My talk is entitled "Discovering Machine Learning."
Ignite talks are the loopy, roller coaster, rides of public speaking. They only last give minutes! Some are very formal demanding 20 slides that auto-advance every fifteen seconds. Bam! A test of creativity and concentration. We didn't talk about those details so I'm packing thirty four slides into my talk. Ha! It's gonna be awesome.
Even if it's a disaster it'll still be awesome! 😜
Supposedly we'll be recorded. Look here for a link to my talk once it's up. I'll add links to my fellow speakers if I can.
The spirit of my talk is introducing a general audience to machine learning. ML is complicated and I find talking about such things easier when I break them down into simpler concepts. Explaining by analogy is helpful too.
Several tech-hero companies have created introductory ML trainings of their own. Each of them has a slightly different angle, but all are well-done. Check them out and see if they help you get a sense of this fascinating tech. ML isn't easy, but it's well worth the effort!
ML is an exciting for opportunity for businesses to leverage all of the data they've been collecting and storing over the past decades. ML is fundamentally about finding patterns in data that reveal valuable business insights.
Many key industries are dominated by a few big players, and most come down to just two choices. In some ways machine learning is the same. There are many open source libraries that address certain scenarios or solve speicifc problems. Ultimately there are two players worth looking into:
Now that I'm actively playing with ML I can see it's an entirely new way of programming computers. It's less about writing code and more about carefully curating data that results in code.
Here's a preview of my deck straight out of Keynote. I'm totally stoked about it!
I've slowly been studying machine learning evenings and weekends over the past five years. Never steady because it's intense, and I want to learn other subjects as well.
When I first looked at ML I thought it was going to be impossible! It looked like what people did was go to a prestigious university and earn a PhD in AI to begin their journey. I didn't have time for that! Or any interest in pursuing another degree.
I decided to pick up another subject from my learning list and dive into it with more dedication. I throttled down on learning ML and put it it into maintenance mode. I kept up by reading blog articles, following practitioners on Twitter, and reading a couple of books each year.
The past few years have been spectacular for machine learning. It seems to me that all of those AI PhD cranked out of uni have been vacuumed up by major tech companies and put to work. It's as if they've turned themselves into working software that all of use can casually use.
Casually for programmers like me who benefit from using libraries, and SDKs, and frameworks. That's great!
They've also come up with a way to democratize it. "Automated" machine learning solutions are provided by the top cloud platform companies. Any company making digital products and services, who don't have data-scientists and AI engineers can still compete using AutoML.
Here are the two tech titans offering tools like these:
If you’re drawn to learning new tech tools, and think machine learning is compelling, try applying it to your chosen field of study. The big platform companies like Amazon Web Services and Google Cloud Platform are rapidly developing easy-to-use offerings that democratize machine learning. You might find ways to deliver previously unsolvable problems in the name of a fantastic user experience.
Please use this presentation as a jumping-off point for your continued learning. You can believe me and what I’ve written here, but go dig into the subject more, and confirm it for yourself.
Replace what I’ve told you with your own findings - that’s totally natural! Then come back and share with us what you’ve discovered when you can.
Special thanks to the amazing Autumn Hood for inviting me to participate in her Ignite event. I always support people who want to build community and I think she's terrific at pulling people together.
I think people want to share some time together and this is a great way to help that happen.