What do you do when you have a mountain of money and time on your hands? You become a patron of the arts. Every industrialist worth his salt knows that, right? So why should the titans of the digital age be any different? The chief difference in this case is that in the case of virtual philanthropy, the recipient of all that largesse is not so much an artiste as AI artiste in training. For the past few years, Google has been gaga about artificial intelligence. So much so that its search engine is powered by it. More importantly, Google has been experimenting with AI to see if machines can make art.
Can you say PC Picasso?
It all started when the world's most popular search engine tried to teach a machine to understand images. An early version of Google's RankBrain was tasked with trying to understand the images.it encountered. Hoping to find a workaround for one of the biggest lapses in SEO, one that makes search engine spiders blind to images and videos, the attempt resulted in hilarity when Google's AI began to morph images it tried to make sense of.
 |
| Image courtesy of iflscience.com |
For example: When asked to display a picture of a dumbbell, it created a representation of the dumbbell, complete with the arm holding it. (See image at right)
Other than answering Isaac Asimov’s query as to whether Androids Dream of Electric Sheep, what Google may have inadvertently done was to create the world’s first PC Picasso. What they also did was digitize the clouds cumulous
It's a poet that doesn't know it
You have to understand that true artificial intelligence can't be programmed. The machine has to learn about the world around it. In order to teach their AI how to form sentences more like a typical human being, researchers fed Google's neural net close to 11,000 novels, including those of the romance and fantasy genre. Next, researchers provided the AI with two sentences, tasking it to generate a response that showed the logical progression from the first statement to the second.
As happened when tasked with a visual interpretation, Google's AI did not provide the expected response. Instead, it took a more roundabout approach that resulted in something quite similar to poetry.
When asked to produce a response from the following two statements:
1. there was no one else in the world
2 I turned to him
This is what resulted:
there is no one else in the world.
there is no one else in sight.
they were the only ones who mattered.
they were the only ones left.
he had to be with me.
she had to be with him.
I had to do this.
I wanted to kill him.
I started to cry.
I turned to him.
there is no one else in sight.
they were the only ones who mattered.
they were the only ones left.
he had to be with me.
she had to be with him.
I had to do this.
I wanted to kill him.
I started to cry.
I turned to him.
Here's another example:
this was the only way.
it was the only way.
it was her turn to blink.
it was hard to tell.
it was time to move on.
he had to do it again.
they all looked at each other.
they all turned to look back.
they both turned to face him.
they both turned and walked away.
it was the only way.
it was her turn to blink.
it was hard to tell.
it was time to move on.
he had to do it again.
they all looked at each other.
they all turned to look back.
they both turned to face him.
they both turned and walked away.
Once again, instead of throwing in the towel, Google's research team became enthralled with the idea of spontaneous machine artistry. So much so that their latest development, named Project Magenta is attempting to teach a machine to play the piano.
Google's blog states that:
Magenta has two goals. First, it’s a research project to advance the state of the art in machine intelligence for music and art generation. Machine learning has already been used extensively to understand content, as in speech recognition or translation. With Magenta, we want to explore the other side—developing algorithms that can learn how to generate art and music, potentially creating compelling and artistic content on their own.
To that end, Google's Brain Team has developed open source code that artists, musicians, researchers and anyone else can access. Google's stated goal is to "build a community where the right people are their to help out." Just like a doting parent posting a classified ad looking for a music teacher to tutor their child, Google's team has put out the word that artists of all stripe are encouraged to participate in Project Magenta. For more information, including demos, blogs and technical papers, go to http://magenta.tensorflow.org/welcome-to-magenta
This ain't your granddad's player piano
 |
| Image courtesy of commons.wikimedia.org |
While early results of Project Magenta aren't anything to give Herbie Hancock a run for his musical money, Google's AI has most definitely made the scene. If you want to hear one of the tunes being cranked out by Project Magenta, click on the link below.
Just as chess playing computers didn't cause grand masters to lose any sleep until IBM's Deep Blue computer beat Gary Kasparov in 1997, machine-based music while still in its infancy could some day compete with humanity on the stage at Carnegie Hall.
 |
| Get your FREE e-Book |
If you'd like a free copy of our eBook, "Internet Marketing Tips for the 21st Century," please fill in the form below and we will give you immediate access to it. Your information is always kept private and is never sold.
Carl Weiss is president of WorkingtheWebtoWin.com a digital marketing agency in Jacksonville, Florida that routinely works with bloggers and other online marketers to grow their businesses.
No comments:
Post a Comment