What is Machine Learning?5:06 with Nick Pettit
Machine learning encompasses many different ideas, programming languages, frameworks, and approaches to the subject, so the term "machine learning" is difficult to define in just a sentence or two. But essentially, machine learning is giving a computer the ability to write its own rules and learn about new things, on its own.
More on Ethics
Check out the Treehouse course on Ethical Design. Stage 3, in particular, covers machine learning.
What happens when a machine intelligence is capable of creating an even better machine intelligence, and so on and so forth? By giving up more and more control to machines, how will humans stay relevant in the workforce? And could a machine intelligence be used as a geopolitical weapon in cyber warfare?
It might seem like science fiction right now, but the pace of advancement is staggering. Technological leaps that were still thought to be decades away are already happening. Just like computers, the Internet, or even splitting the atom, this is a powerful new area that's capable of both great and terrible things, depending on how we decide to use it.
It's also possible for unintentional bias to make its way into a machine learning model. For example, in the tech industry, there are many issues of bias around under representation and pay disparity. This same mindset can create biased datasets that are then reflected in the outcomes of machine learning models.
Here are some additional perspectives from key figures in the tech industry:
[MUSIC] 0:00 Hi I'm Nick. 0:09 In this course we're going to learn about the broad topic area of machine learning. 0:10 We'll explore some of the big ideas and, toward the end, 0:16 we'll even write a little bit of code in Python. 0:19 I can make some intelligent predictions. 0:22 If you don't know Python already, 0:24 I highly recommend you start with some of the prerequisites for this course. 0:26 You'll need to install Anaconda, 0:31 which is a Python based platform focused on data science and machine learning. 0:33 We won't install Anaconda in these videos. 0:39 So if you haven't installed it already, 0:42 check out the notes associated with this video for 0:44 help on how to do that, along with some other Python resources on Treehouse. 0:47 Machine learning encompasses many different ideas, programming languages, 0:53 frameworks and approaches to the subject. 0:58 So the term machine learning is difficult to define in just a sentence or two. 1:00 But essentially, machine learning is giving a computer the ability 1:07 to write its own rules and learn about new things on its own. 1:13 In this course, you're going to hear a lot of new terms and 1:18 ideas you may not be familiar with. 1:22 It's a lot to take in and you might not understand it all right away. 1:25 That's normal. 1:29 I was confused the first time I learned about these concepts. 1:30 I encourage you to take a look at the teacher notes after each video and 1:33 watch videos more than once if you feel like you need to. 1:37 That said, machine learning actually is not as difficult as you might believe. 1:42 And its applications are far reaching. 1:47 You might be surprised where machine learning could show up in your life and 1:50 how it might be useful to you now and in the coming years. 1:55 Just to name a few applications, 1:58 machine learning is already unlocking new possibilities like self-driving cars, 2:00 digital assistants and chat bots, and new approaches to science and health care that 2:06 can help us dig through vast libraries of complicated information in an instance. 2:11 Machine learning can be used to create spam filters, search engines that can find 2:17 your photos based a text description, and even create original fine art and music. 2:22 It's putting the power of human reasoning into a computer. 2:28 In the past, this topic has had other labels like artificial intelligence, 2:32 and that term still applies here. 2:38 What's different, however, is that past approaches to machine learning 2:40 involved a human being writing most or all of the rules. 2:45 In the mid 90s, the IBM Deep Blue super computer famously won a series of 2:49 chess matches against Garry Kasparov, 2:54 who many considered to be one of the greatest chess players ever. 2:56 This is an impressive accomplishment even by today's standards. 3:00 However, older approaches to artificial 3:04 intelligence often involved millions of lines of code. 3:07 Writing if statements and conditional logic to guide the computer through 3:10 an incredible number of small decisions. 3:15 I'm oversimplifying, but in essence Deep Blue would look at many 3:18 branching chess moves into the future, far more than a human can do. 3:23 And it would pick the best one based on a complicated set of predefined criteria. 3:27 This type of approach only had the appearance of intelligence. 3:34 When in reality, 3:37 most AIs could be distilled down to a giant flow chart created by people. 3:38 And more importantly, Deep Blue could only play chess. 3:44 Ultimately, it's more desirable to create a generic machine intelligence 3:48 that can perform a variety of tasks. 3:53 Put simply, machine learning is still artificial intelligence but 3:56 with a strong focus on the computer's ability to write its own set of rules as 4:00 it processes more data. 4:05 Rather than humans teaching it everything upfront. 4:06 Using machine learning tools and algorithms, 4:10 we can explore data in new ways that would otherwise take a ton of work by humans. 4:13 Or even by traditional programming or data processing approaches. 4:17 There's one more thing I need to mention. 4:22 While this technology has the potential to be more impactful than even 4:24 the microprocessor or the Internet. 4:29 It also introduces many ethical questions that, over the next few years and 4:31 decades, will need answers from people like you. 4:36 This is a huge ongoing discussion on its own and 4:40 it's outside the scope of this course. 4:43 But I highly recommend you check the notes associated with this video for 4:45 more resources and insight into the ethical and 4:51 societal implications of machine learning. 4:54 Now let's continue with this broad overview by taking a closer look at some 4:57 of the different approaches to machine learning. 5:02
You need to sign up for Treehouse in order to download course files.Sign up