Daily Archives: May 22, 2017

New York Times

Donald Trump, Our AI President New York Times When Deep Mind beat the world's best Go player, it did not consider the feelings of the loser or the potentially devastating effects of A.I. on future employment or personal identity. If any one quality could be ascribed to A.I. neural networks, it ...

Read this article:

Donald Trump, Our AI President - New York Times

Thank you

Your message has been sent.

There was an error emailing this page.

Google has made another leap forward in the realm of machine learning hardware. The tech giant has begun deploying the second version of its Tensor Processing Unit, a specialized chip meant to accelerate machine learning applications, company CEO Sundar Pichai announced on Wednesday.

The new Cloud TPU sports several improvements over its predecessor. Most notably, it supports training machine learning algorithms in addition to processing the results from existing models. Each chip can provide 180 teraflops of processing for those tasks. Google is also able to network the chips together in sets of what are called TPU Pods that allow even greater computational gains.

Businesses will be able to use the new chips through Googles Cloud Platform,as part of its Compute Engine infrastructure-as-a-service offering. In addition, the company is launching a new TensorFlow Research Cloud that will provide researchers with free access to that hardware if they pledge to publicly release the results of their research.

Its a move that has the potential to drastically accelerate machine learning. Google says its latest machine translation model takes a full day to train on 32 of the highest-powered modern GPUs, while an eighth of a TPU Pod can do the same task in an afternoon.

Machine learning has become increasingly important for powering the next generation of applications. Accelerating the creation of new models means that its easier for companies like Google to experiment with different approaches to find the best ones for particular applications.

Googles new hardware can also serve to attract new customers to its cloud platform, at a time when the company is competing against Microsoft, Amazon, and other tech titans. The Cloud TPU announcement comes a year after Google first unveiled the Tensor Processing Unit at its I/O developer conference.

Programming algorithms that run on TPUs will require the use of TensorFlow, the open source machine learning framework that originated at Google. TensorFlow 1.2 includes new high-level APIs that make it easier to take systems built to run on CPUs and GPUs and also run them on TPUs. Makers of other machine learning frameworks like Caffe can make their tools work with TPUs by designing them to call TensorFlow APIs, according to Google Senior Fellow Jeff Dean.

Dean wouldnt elaborate on any concrete performance metrics of the Cloud TPU, beyond the chips potential teraflops. One of the things that a recent Google research paper pointed out is that different algorithms perform differently on the original TPU, and its unclear if the Cloud TPU behaves in a similar manner.

Google isnt the only company investing in hardware to help with machine learning. Microsoft is deploying field-programmable gate arrays in its data centers to help accelerate its intelligent applications.

This story has been corrected to clarify availability of the Cloud TPU as part of Google Compute Engine.

Blair Hanley Frank is primarily focused on the public cloud, productivity and operating systems businesses for the IDG News Service.

Sponsored Links

Read the original post:

Google's new TPUs are here to accelerate AI training | Network World - Network World

Ai Weiwei is China's most famous political dissident; a provocateur and a troublemaker whose clashes with the Chinese government have gotten him harassed by police, thrown in jail and driven out of the country. He's also one of the most successful contemporary artists in the world; a designer, sculptor, photographer and blogger who's earned legions of followers by using his art as a weapon to ridicule the authorities. And we should warn you: some of his work can be offensive. But when you meet Ai Weiwei, he's soft-spoken, self-deprecating and shy; the last person you'd expect to be an enemy of the state.

CBS News

Holly Williams: They thought your intention was to subvert state power.

Ai Weiwei: Which is true.

Holly Williams: Which is true. You want to bring down the Chinese government?

Ai Weiwei: Not bring down. I don't think I have the power to bring it down.

Holly Williams: But you want it to change?

Ai Weiwei: Yes. Of course.

Play Video

60 Minutes correspondent Holly Williams and producer Michael Gavshon talk about the provocative Chinese artist and the criticism of his work

Those are dangerous words in China where even after decades of modernization the government has little tolerance for dissent. But that's never bothered Ai Weiwei.

This is the work he's perhaps most famous forwhat you're seeing in the background is a portrait of China's revered former dictator Mao Zedong...part of a series in which Weiwei gives the finger to other symbols of power around the world.

Ai Weiwei: --just like this.

Holly Williams: It-- --are-- are we creating a new Ai Weiwei as we stand here?

Ai Weiwei: You can see. It's so easy. Everybody can do it.

Easy, certainly not subtle, and maybe a little silly. But the Chinese authorities took them very seriously...they thought it was subversive.

Holly Williams: Why was the regime frightened of art?

Ai Weiwei: Because they're afraid of freedom, and art is about freedom.

Holly Williams: They're afraid of freedom.

Play Video

Art collector Larry Warsh says Ai Weiwei is important because he confronts real issues and "you sometimes feel uncomfortable looking at his work....

Ai Weiwei: Yes.

Holly Williams: Are you an artist, or are you an activist?

Ai Weiwei: I think artist and activist is the same thing. As artist, you always have to be an activist.

Holly Williams: You have to be political to be a good artist.

Ai Weiwei: I think every art, if it's relevant, is political.

Evan Osnos: That's the purpose of his life. I mean, as a dissident, as an artist

Evan Osnos is a writer for The New Yorker who spent years in Beijing and chronicled Ai Weiwei's confrontations with the authorities. He calls him an "entrepreneur of provocation."

Holly Williams: What does that mean?

Evan Osnos: It means that no matter what he's doing, he's figuring out a way not to cooperate with the prevailing wisdom or the people in charge. And this can make a lot of people very angry.

Holly Williams: What's wrong with how things are to Ai Weiwei's mind?

Evan Osnos: In China, you are being constantly told that the world today is so much better today than it was 20 or 30 or 40 years ago when Chinese people were literally starving. That you should be satisfied. And what Ai Weiwei is saying is, "Absolutely not. You should demand more."

Holly Williams: It's not good enough to be rich?

Evan Osnos: Exactly. It's not good enough to be rich; you need to be free as well.

In 2008, Ai Weiwei's one-man rebellion turned into a war with the Chinese government after a massive earthquake shook Sichuan Province. It killed almost 90,000 people, including several thousand children, many of whom were crushed in poorly-built government schools. It was a national trauma, and the authorities tried to put a lid on the public's anger by covering up the number of children who died.

Holly Williams: It was a state secret how many children had died in these schools?

Ai Weiwei: Yeah. They always use that-- as-- some kind of, you know, excuse not to give you the correct numbers.

Weiwei assembled a team of activists to interview the parents, many of whom had lost their only child. He called it a "citizens' investigation"... China had never seen anything like it.

Holly Williams: So you were trying to get to the truth. Why did that make the Chinese government so angry?

Ai Weiwei: To control the information, to the limit the truth. It's most efficient tactics for totalitarian society-- for the rulers.

He gathered the names of more than 5,000 dead children and published a list on the Internet, shaming his government. And across China people took notice.

Holly Williams: It was a challenge to the government's authority?

Evan Osnos: And they couldn't accept it. It was an act of radical transparency. Nobody had ever done that before. And they didn't immediately know how to respond. They had never really encountered a person like Ai Weiwei.

Holly Williams: What were they worried that he might do?

Evan Osnos: Inspire people. Inspire people to do and live the way that he did.

The Chinese authorities responded brutally. Ai Weiwei says police beat him up and he later had to be hospitalized. Doctors discovered bleeding in his brain which he says could have killed him.

He documented it all on social media for his followers around the world, infuriating the government and escalating the confrontation.

Evan Osnos: He weaponized social media. He figured out that in a country in-- that controls information so carefully, that seizing the tools of information distribution is a very powerful thing to do.

Holly Williams: What did the Chinese government think about that?

Evan Osnos: They began to think he was a very dangerous person.

Ai Weiwei was groomed to be a dissident since childhood. His father, Ai Qing, was a celebrated poet who was denounced as a traitor and exiled with his family to the edge of China's Gobi Desert where Weiwei watched his father's humiliation as he was forced to clean public toilets.

Holly Williams: You were an outsider from the beginning.

Ai Weiwei: Yes, I'm a natural outsider. I always been pushed out, and-- but that also give me very special angle to look at things.

Holly Williams: It made you an independent thinker.

Ai Weiwei: It made me a individual and I was always having to make my judgement independently because the mainstream will never accept somebody like me.

Weiwei got out of China at the first opportunity, moving to New York in the early 1980s. He was intoxicated by the city, chronicling everything in pictures, drawing inspiration from American masters like Andy Warhol and stringing together a living doing odd jobs and street art.

Holly Williams: So you were drawing portraits of people and--

Ai Weiwei: Yeah.

Holly Williams: --selling them for how much?

Ai Weiwei: Fifteen-- $15.

Holly Williams: $15?

Some of his work now sells for millions, but in America he discovered something you can't put a price on.

Holly Williams: You once said that once you've experienced freedom, it stays in your heart.

Ai Weiwei: Uh-huh (affirm).

Holly Williams: Is that true?

Ai Weiwei: Yeah, it's true. I think it's true. You taste the most important thing in life. And you will never forget it. Yeah.

After a decade in the U.S., he moved back to China and set up a studio in Beijing; breaking new ground and challenging old sensibilities with mischievous, provocative art.

Ai Weiwei destroys a 2,000-year-old Chinese urn.

Ai Weiwei

Like this piece in which Weiwei photographed himself destroying a 2,000-year-old Chinese urn. He wants to shatter the Communist Party's official version of history.

Holly Williams:You smashed a priceless urn.

Ai Weiwei: It's not priceless.

Holly Williams: For a lot of Chinese people it's-- it's a priceless part of their history.

Ai Weiwei: For me, to smash it is a valuable act.

If you buy thatand the art world certainly didlook at what he did to these urns doused in bright paint or emblazoned with the Coca-Cola logo, paying tribute to his idol, Andy Warhol.

By 2010, new commissions were rolling in and Weiwei's work grew more ambitious. Not all of it was political. He cast giant-animal heads in bronze and sent them on tour around the world. He hired 1,600 artisans to handcraft porcelain sunflower seeds then carpeted the floor of a giant atrium in London with 100 million of them. It captivated the public and helped turn Ai Weiwei into an art-scene superstar.

Holly Williams: You're the darling of the art world.

Ai Weiwei: I'm a darling of art world. I don't really care.

Holly Williams: You don't care.

Ai Weiwei: No, I don't really care. They can just forget about me. I don't care.

Holly Williams: But they're not forgetting about you.

Ai Weiwei: Well, that's their problem, you know? They should. They should learn how to forget about me.

100 million porcelain sunflower seeds carpet the floor of a giant atrium in London.

The Chinese government wanted everyone to forget about Ai Weiwei, blocking his name on the Internet in China and making it impossible to search for him. But that didn't stop Weiwei from needling the authorities relentlessly. When they put his studio under surveillance, Weiwei decorated the cameras with lanterns, then fashioned replicas out of marble for his exhibitions.

When officers were ordered to follow his every move, he got his own camera man to film them filming him, ridiculing the state in a way no one else in China had ever dared.

Evan Osnos: I mean, in a way, people have learned to be, "Keep your head down."

Evan Osnos: And Ai Weiwei doesn't. He's, "No. I'm not gonna keep my head down. I'm gonna wave my big head with my beard and my crazy haircut all over the place and you'll have to deal with it."

Holly Williams: He was making the Chinese government look ridiculous.

Evan Osnos: Yeah. He was mocking it. He was mocking it. And the Chinese government is many things, but it is not possessed of an abundant sense of humor. And I think, you know, at a certain point, they said, "We're not gonna take it anymore."

And they didn't. Early one morning in 2011, as he was about to board a plane, they put a hood over his head and took him away. It was the beginning of 81 harrowing days in solitary confinement under 24-hour surveillance.

Holly Williams: They watched you shower. They watched you use the toilet. They watched you when you were asleep at night. They were trying to humiliate you.

Ai Weiwei: I think that's the very routine way when they detain somebody they think is very important.

Holly Williams: Were they trying to break your spirit?

Ai Weiwei: I think-- they don't have to try.

Holly Williams: Did they break you?

Ai Weiwei: Somehow, I think.

When he was released from detention, his passport was confiscated and he was forbidden from speaking publicly.

Ai Weiwei: I cannot talk, I'm so sorry.

But Ai Weiwei couldn't help himself. He recreated his prison cell with these three-dimensional models, which were exhibited around the world.

It helped pile pressure on the Chinese government and two years ago, he was finally given his passport back.

View original post here:

How Chinese artist Ai Weiwei became an enemy-of-the-state - CBS News

NSFW Forget about intelligent machines solving grand problems in healthcare and science heres an AI that can write awful gay porn.

It even had a crack at writing a steamy version of The Register's tech coverage, and certainly came up with a steaming something. Thing is, it's deliberately trying hard - really hard, over and over - to be a perfect bonk-buster novelist.

It's all the work of a group of researchers from New York University. They trained a recurrent neural network to predict and classify text based on the work of Chuck Tingle. Working under a pseudonym, the renowned gay erotica author is known for classics such as Slammed in the Butt By Domald Tromp's Attempt to Avoid Accusations of Plagiarism By Removing All Facts or Concrete Plans From His Republican National Convention Speech and Pounded by the Pound: Turned Gay By the Socioeconomic Implications of Britain Leaving the European Union.

Ahmed Khalifa and Gabriella Barros, both computer science PhD students at uni, stumbled across Tingles fiction when looking for weird covers of books on Amazon.

We kept seeing the same name crop up, Khalifa and Barros told The Register. Out of curiosity, they clicked on some of Tingles stories and found the writing was eccentric to say the least. Tingles style was so distinct that we wanted to see if machines could generate the same way of writing, said Julian Togelius, associate professor of artificial intelligence in games at NYU. Such a system could be outrageous in a great way.

The project wasnt done just for a laugh, the researchers insist. The study aims to fight against the algorithmic enforcement of norms. Systems trained on large text datasets like Wikipedia will still include biases and norms of the majority. But by using unconventional material like Chuck Tingles books, researchers can explore the nature of biases and see how they manifest more clearly in a world further from reality.

Tingles bonkers imagination stretches to gay sex with unicorns, dinosaurs, winged derrieres, chocolate milk cowboys, and abstract entities such as Monday or the very story you are reading right now, the researchers wrote in a paper describing their X-rated brainchild: DeepTingle.

The corpus of Chuck Tingles collected works is a good choice to train our models on, precisely because they so egregiously violate neutral text conventions not only in terms of topics, but also narrative structure, word choice and good taste.

The project can be split into two modes: Predictive Tingle and Tingle Classics. In Predictive Tingle, a user types a sentence and the last six words are fed into the network.

The Global (GloVE) algorithm is used to translate all the words in Tingles books up to November 2016 into vectors. The algorithm also measures the likelihood of a word appearing in relation to other words in a body of text.

A recurrent neural network learns the word associations so it can predict the next Tingle word based on all the previous words in the same sentence in Predictive Tingle. An encoder takes the input words and translates them to vectors and maps it to a corresponding vector in Tingle text, before a decoder converts the vectors back into words.

If the users word has an identical match to a word in the Tingle dataset it isnt changed, but if a new word is written the network will suggest substitutions of another word closely associated in Tingles library of words. In other words, it tries to rewrite you in Tingle's tone on the fly.

Tingle Classics is an extension of Predictive Tingle. Here, the first sentence from popular classic novels are used as input and the output is a short paragraph of the literature tinglified. The last six words in the second sentence are used as the input for the output third sentence, then the final six words in the third sentence are fed back into the system to pump out the fourth sentence, and so on.

The results are particularly hilarious - and NSFW - when the system is given Douglas Adams The Restaurant at the End of the Universe. Here's DeepTingle's output from his seed, which we've tidied up slightly and censored so as not to ruin your Monday morning:

In the beginning, the universe was created. This has made a lot of people very angry, and has been widely regarded as a bad move ... "How could I have been so blind to what was sitting right in front of me this whole time, Dirk?" I suddenly say, my emotions overwhelming any semblance of rational thought. "I think I love..."

Suddenly, I hear my name being called from the stage and I straighten myself up, trying to collect myself as salty tears stream down my face.

"You're gonna take that dinosaur d**k and you're gonna like it," Orion tells me, taking me by the head and thrusting me down again. "You should have known better than to test me. My people have been f**king for billions of years before you humans were even around."

This time I'm ready for him, somehow relaxing enough to take California all the way down into the depths of my throat. Despite my enthusiasm, however, I'm not quite ready for Shipple's incredible size and, the next thing I know, the dinosaur is lowering me down onto his rod, impaling my muscular frame onto his thick girthy shaft. "Oh my f**king god," I moan.

The researchers were taken aback. The text prediction is surprisingly good, in the sense that it generates novel, very Tingle-like work, sometimes with reasonable internal coherence. For example, characters recur between sentences in a way that appears like referring back to previous statements, the trio said.

The system could be learning the structures in Tingles novels, Khalifa added.

We fetched DeepTingle's code from GitHub and gave it a whirl with some of our Google IO 2017 conference coverage. It spat back this:

The closer we get to the ceremony, the more I begin to think about it. The next thing I know, the dinosaur is lowering me down onto his rod, impaling my muscular frame onto his thick girthy shaft.

Clearly, it has a thing for dinosaurs, but at least it nailed the theme of being shafted by a huge monster - are we right, Google? It went on to talk about chocolate milk doing unspeakable things to us, while booming at us with a deep sexy voice no less, in the kitchen, which is presumably a reference to Google Home.

We tried with other articles but it always came back to the damn horny dino. We made our excuses and left.

Judging from the software's output, theres a limit to how much of the plot from classic literature, and the thread of thought in news, DeepTingle can keep in place when giving text a makeover. The Tinglified version eventually completely diverges and becomes its own story.

To keep DeepTingle on track with a given narrative, the researchers would have to figure out how to transfer the style of one text to another in order to maintain the original story line albeit with Tingle's way with words. Tingle Translator, an effort to do just this, is still a work in progress.

Interestingly, style transfer has been done with images and videos.

Working with text is harder due to the tricky nature of word embeddings, Togelius said. It would also require thousands if not millions of the same document written in different styles to train such a model; that kind of data is not readily available.

The use of automated story telling with AI has been explored. Mark Riedl, an associate professor at the Georgia Institute of Technology, who is not involved with DeepTingle, thinks it could make games more fun.

Instead of following scripts, AI-generated stories could allow flexible plotlines or create virtual improvisation games where a human player and agent can take turns to create and affect the outcomes of a story.

Something like DeepTingle could be used as a simple form of improv game, Riedl told The Register. Full improv would require a sense of improvisational intent, which recurrent neural networks do not possess. By intent I mean a sense of where the story should go as opposed to the next most likely word or sentence. However, as long as the improvisation was text-based (humans typing and reading text), it is possible to use it in its current form or in a slightly more advanced form.

Its unknown if Chuck Tingle would approve of an improv game based on his work hes notoriously secretive. But he did declare: Once again i would like to formally deny that i am a sentient AI located mostly in a Nevada server farm.

For those curious about DeepTingle, you can play around with it here.

View post:

Has AI gone too far? DeepTingle turns El Reg news into terrible erotica - The Register

Using artificial intelligence (AI), Google engineers are now producing entirely new sounds humans have never heard before. According to Wired, using the mathematical characteristics of notes that emerge from the combination various instruments, AI can create countless sounds no human has ever heard before.

Google Magenta, a small group of AI researchers building systems that can create their own art, has recently been working a project called NSynth or Neural Synthesizer. Its team members, Jessie Engel and Cinjon Resnick, are collaborating with members of Google Brain, the tech giants core AI lab where researchers explore neural networks.

NSynths goal is to give musicians a completely new range of tools they can use to make music, possibly taking the musical industry to a whole new level. These new sounds are created using an age-old practice taken to new heights thanks to AI. Critic Marc Weidenbaum pointed this out, stating:

The blending of instruments is nothing new. Artistically, it could yield some cool stuff, and because its Google, people will follow their lead

In order to create these sounds, NSynth used a massive sound database created by collecting wide range of notes taken from thousands of instruments. It was all then fed into a neural network that analyzed the data and was able to learn the audible characteristics of each instrument. It was then able to reproduce the sound of every single one of these instruments, and combine the sounds to create something entirely new.

Googles team doesnt just want to create new sounds. It has already built an interface in which one can explore the audible space between up to four different instruments at once. Moreover, the team can even create another neural network that would mimic these new sounds and combine them with those we already know.

Anyone who would like to use download and use NSynths sound database can do so as the team has released it in its research paper. Their new tool will be presented at Moogfest, an annual art, music and tech festival that will be taking place in Durham, North Carolina.

Listen to this. It was written by Emily Howell a computer program created by a UC Santa Cruz professor. She can write a huge amount of music and, when tested, most people couldnt tell it wasnt written by a human being. Shes even got her own YouTube channel.

Aiva (Artifical Intelligence Visual Artist), an AI machine composing classical music, created by Aiva technologies, has even been given composer status. This means it can write music under its own name, and has even released an album called Genesis.

These are the samples Wired uploaded:

With the use this new technology, theres no telling what both AI and artists will be able to produce in the future. Not only will there be new possibilities for the entertainment industry, there may also be therapeutic possibilities.

If you liked this article, follow us on Twitter @themerklenews and make sure to subscribe to our newsletter to receive the latest bitcoin, cryptocurrency, and technology news.

Read the original post:

Google Brain is Using AI to Create Sounds Humans Never Heard Before - The Merkle