15.04.2018 Views

programming-for-dummies

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

652<br />

Machine Learning<br />

Such image recognition software examines the shape of a person’s face along<br />

with the physical position of the nose in relation to the eyebrows, lips, and<br />

chin. No matter how many disguises cheaters may wear, they can’t hide or<br />

change the physical dimensions of their faces, which is how such image<br />

recognition programs can spot them.<br />

In the field of robotics, image processing is crucial because it gives robots<br />

the ability to sense their environment and avoid or move toward certain<br />

objects. Teaching a robot to avoid objects in its path might seem easy until<br />

the robot looks through a plate glass window and mistakes the window <strong>for</strong><br />

an opening that it can roll through to get to the other side.<br />

Although image recognition might help a robot identify the edges of walls or<br />

the end of a road, image recognition must also teach a computer to recognize<br />

shadows or reflections that can distort images. Primitive image recognition<br />

might simply distinguish between patches of light and dark, but more<br />

sophisticated image recognition could not only recognize an object in its<br />

path but also identify what that object might be, which involves a related<br />

category of artificial intelligence known as image processing.<br />

In a limited domain of objects, seeing and understanding images can be determined<br />

in advance, but in the real world, the number of images a computer<br />

might see and need to recognize is nearly infinite. To solve the problem of not<br />

only recognizing an image but also understanding what that image might be,<br />

computers need to learn independent of their human programmers.<br />

Machine Learning<br />

The preceding examples of problem-solving artificial intelligence programs<br />

only mimic intelligence, but all a computer’s intelligence is programmed by<br />

humans ahead of time. When faced with different situations, these problemsolving<br />

programs behave the same way no matter what the circumstances<br />

might be.<br />

That’s why the second, and more important, focus of artificial intelligence<br />

involves teaching computers to learn, otherwise known as machine learning.<br />

Machine learning can involve training by humans, but it’s more commonly<br />

associated with self-learning in an unsupervised environment.<br />

One way to mimic machine learning is to insert problem-solving knowledge<br />

into a program and then allow the program to modify its database of knowledge.<br />

That’s the idea behind two popular <strong>programming</strong> languages — LISP<br />

and Prolog — which are specially designed <strong>for</strong> artificial intelligence.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!