Machine Learning: Exploring New Forms of Artificial Intelligence

On Oct. 26, the Mount welcomed Mark Shipley for a presentation on machine learning in Laughlin Auditorium.  This presentation titled “Machine Learning and Deep Learning: If I don’t even know what it is, how can I already be behind and why should I care?” was part of the ACM/MAA lecture series here at the Mount.  ACM stands for the Association of Computing Machinery and is one of the organizations on campus.  Students in math and science departments could receive two professional development points for attending the hour-long talk.

Shipley is a former graduate of the Mount in ‘98 with a degree in computer science and mathematics.  While at the Mount, he was one of the first student webmasters for the Mount’s website.  Shipley spent most of his career as an executive consultant at CGI, an IT consulting firm, before working as a director for CDG, a Boeing Company.  He just recently assumed the position of Vice President at CDG.

The first part of Shipley’s talk was about what the areas of artificial intelligence, machine learning and deep learning were all about.  Shipley defined artificial intelligence using Forbes as “the broader concept of machines being able to carry out tasks in a way that we would consider ‘smart.’”  In other words, artificial intelligence is a way of programming machines to act like humans.

Now not every machine that acts like humans are considered artificial intelligence. An amazon echo is a form of artificial intelligence, while a Roomba robotic vacuum is not.  Forbes, defines machine learning as “a current application of AI based around the idea that we should really just be able to give machines access to data and let them learn for themselves.”

Machine learning is exactly what it sounds like, the ability for machines to learn on their own through programming.  Machine learning allows machines to reach a level of programming beyond what a human could produce.  Thousands of lines of code could not achieve the level of programming machine learning can achieve.  For a machine to “learn”, it must be given vast amounts of data to analyze.  According to Forbes, machine learning can analyze writing and determine whether that writing is a complaint or congratulation.

Deep learning goes beyond machine learning in that not only can it recognize complex patterns, but it can also predict results.  Deep learning is the least developed of the three areas of artificial intelligence.  Some examples of the application of deep learning include Tesla self-driving cars (analyzing billions of incidents to teach cars to drive), NBA player analysis (why is Steph Curry off for one game?) and areas of healthcare (predicting cancer in patients).

And yet, as Shipley says, deep learning is failing; it is not fully functional.  We do not have fully self-driving cars, we cannot even predict weather very well.  People are pouring money into deep learning, Shipley says, and it is currently a very lucrative field.  But the progression of deep learning is exponential not linear, according to Shipley.  All this seemingly futile effort will suddenly pay off and it is a question of who will be there when it does.

While the results of machine and deep learning have not come to fruition, the efforts behind them are always improving.  Companies are learning as they advance and are making progress in these fields.  Machine learning and deep learning have the potential to revolutionize the modern world.  While it may seem like a waste of time at the moment, it is important to be aware of these concepts should they ever pay off.  “This is the future,” Shipley says, “if we don’t do it we will be left behind.”

Leave a Reply