Skip to Content

Machine Learning with Amazon SageMaker

Posted on October 25, 2022 by

Categories: AWS

Tags:

This post is a synopsis of a lecture I delivered at the yearly Webstep’s “Kompetensbio” event. Every year Web step welcomes all developers to this free event in some tremendous local cinema, where they may enjoy excellent tech talks and watch some thrilling movies. This year, the event was held in Uppsala, Malmö, and Stockholm for the first time.

Introduction

Even since I was a youngster and to this day, I have been a major Science-fiction lover. Growing up in a tiny town in an Eastern bloc country was not actually a lot of fun. Especially if you were educated and interested. Questioning authority was not permitted as well as asking a lot of inquiries.

For a youngster with a vivid imagination and many questions, the only way out was to locate some other worlds, realms where anything was conceivable, and anything was allowed. In the worlds of Arthur Clarke, Isaac Asimov, and others, that is where I felt at home.

Trying to visualize all those faraway planets with advanced nations with superior technology, I could not prevent feeling melancholy because I believed that all that lay in a very distant future. And yet, here we are today, talking about and experiencing some of those technologies: artificial intelligence, self-driving cars, real-time image-recognition systems, etc.

We are indeed living in fascinating times. And these great new technologies provide many opportunities for everyone to join engaged. Like with anything, there is a lot of miss information and noise. For that reason, in the following article, I will try to refute certain fallacies regarding AI and also address the following questions:

  • What is Machine Learning?
  • Why is it important to you, and why should you care?
  • How does it work?
  • When should it be utilized, and when not?
  • How may it be done efficiently?
  • How might AWS SageMaker assist with that?

The Hype

There is a lot of excitement these days about Machine learning, Artificial intelligence, and similar technologies. As usual with the media and reporters, “there is no better news than bad news.” And nothing captures our attention better than terror. That is why writers constantly write out “killer robots” that arwillestroy us all, or at least take all our jobs and leave us homeless on the street.

Honestly, nothing could be further from the truth. Those sophisticated battle machines are far away in the future, assuming they ever happen. I am confident that a lot of legislation and treaties will prohibit something like that from ever happening. Things will change in that sense regarding our work, but not as catastrophic as the media want us to assume. We shall come to that later.

The change

In the previous decade, we have watched firms like Airbnb, Netflix, Amazon, etc., radically alteredaltersectors. Apart from being very successful, all these organizations have something else in common: they leverage data to develop their business models. And all of them have automated much of the business procedures that operate their day-to-day operations.

What we are witnessing now is precisely what Marc Andreesen famously prophesied in his article: Why is software consuming the world? In essence, he wants to imply that someday any process that can be automated will be by putting it into some form of software. And any firm, regardless of its past, will become a software company.

The revolution

These developments are part of a much broader tale that has been going on for centuries. It is a narrative about the Industrial revolution. The one we are in right now is the 4th Industrial revolution. All three previous revolutions started with critical technical inventions: the steam engine in 1760, electricity in 1820, the computer in 1960, and AI and ML (machine learning) (machine learning).

Thinking about this for a time, I concluded that the natural character of industrial revolutions has altered. We shifted from changing raw materials into energy to power businesses — to transforming raw data into insights and information that fuels intelligent services and gadgets. With this, also automation of occupations has altered. So termed “blue collar” occupations were first, now “white collar” jobs are being mechanized.

In the past, every revolution had its winners and losers. This time it is no different. Some anticipate that hundreds of millions of jobs will be destroyed. And the same or even more significant number of new, distinct occupations will be created. Trillions of dollars of profit will be produced each year by those firms that learn to use Machine learning and AI to their advantage.

In my imagination, I imagine this massive transition as a large ocean wave. And we all must question ourselves – where do we want to be when it comes? On the top, surfing and enjoying the time of our lives? Or at the bottom, risking getting crushed by this massive wave? Or on the beach, observing safely from a distance but missing all the fun?

Definitions

What is Machine learning? There are various definitions, but one remark from great hockey player Wayne Gretzky explains that is the best. He was not talking about machine learning when he said: “I skate where the puck is going to go, not where it has been.” He was only addressing the question from the reporter, how come he plays so well? In my perspective, the same logic applies equally to the fundamental essence of machine learning: trying to forecast the next optimal play to win the game.