Building AI is complicated, but understanding it doesn't have to be. Most existing artificial intelligences are just very good guessing machines (like our brain). You feed a bunch of data (such as the numbers 1-10) and ask it to make a pattern (x+1, starting at 0) and make a prediction. (The next number will be eleven.) There is no magic, other than what humans do every day:using what we know to make assumptions about things we don't know.
SummaryArtificial IntelligenceMachine LearningDeep LearningConclusion:AI =Apocalyptic Intelligence?What sets AI apart from other computer programs is that we don't have to program it specifically for every scenario. We can teach it things (machine learning) and it can also teach itself (deep learning). Although there are several varieties of each, they can be broadly defined as follows:
The broadest possible definition of AI is simply that it is a machine that thinks like a human. It can be as simple as following a logical flowchart, or it can be an almost human-like computer that can learn from a wide variety of sensory inputs and apply that knowledge to new situations. That last part is key – the powerful AI everyone imagines is one that can connect all kinds of learned data points to give it the ability to handle almost any situation.
Right now, the AI is still on a pretty narrow track – Alexa is an amazing butler, but she can't pass a Turing test. We currently have a limited form of AI, but it's worth remembering that the definition is so broad that it could possibly cover programs that DeepMind looks like a calculator.
Without machine learning, existing AI would mostly be limited to iterating through long lists of "if x is true, do y, else, do z". This innovation, however, gives computers the power to figure things out without being explicitly programmed. As an example of one type of machine learning, let's say you want a program to be able to identify cats in images:
As complicated as machine learning may seem, it can be boiled down to this:“Humans tell computers what to look for, and computers refine those criteria until they have a model. It's pretty simple, extremely useful, and it filters your spam, recommends your next Netflix shows, and changes your Facebook feed. Try Google's Teaching Machine for a quick hands-on demonstration!
In 2018, it is the cutting edge of AI. Think of it as machine learning with deep "neural networks" that process data much like a human brain. The main difference from its predecessor is that humans don't have to teach a deep learning program about what cats look like. Give him just enough pictures of cats, and he'll figure it out on his own:
To sum it up:deep learning is machine learning where the machine trains, although that's way beyond cats - neural networks are now able to accurately describe everything in an image.
Deep learning requires far more initial data and computational power than machine learning, but it's starting to be deployed by companies from Facebook to Amazon. The most infamous manifestation of machine learning, however, is AlphaGo, a computer that played games of Go against itself until it could accurately predict the best moves well enough to multi-beat. times several world champions.
Hollywood is responsible for a lot of bad science, but when it comes to AI, truth and fiction aren't that far apart. It is not inconceivable that a robot could take control of a space station (2001:A Space Odyssey), make you fall in love (Elle) or behave exactly like a human (Blade Runner, Ex Machina).
That doesn't make it a bad bet, though. AI could accelerate human progress faster than almost anything that had come before it. And, while that may sound cynical, the reality is that if responsible scientists stay away from AI because of its potential for slippage, it will likely be developed anyway. by people with fewer security concerns. We've taken computers from checkers to Go, and the next steps could take humanity to some interesting places.
