Daily Archives: August 14, 2017

As Mr. Eck says, these systems are at least approaching the point still many, many years away when a machine can instantly build a new Beatles song or perhaps trillions of new Beatles songs, each sounding a lot like the music the Beatles themselves recorded, but also a little different. But that end game as much a way of undermining art as creating it is not what he is after. There are so many other paths to explore beyond mere mimicry. The ultimate idea is not to replace artists but to give them tools that allow them to create in entirely new ways.

In the 1990s, at that juke joint in New Mexico, Mr. Eck combined Johnny Rotten and Johnny Cash. Now, he is building software that does much the same thing. Using neural networks, he and his team are crossbreeding sounds from very different instruments say, a bassoon and a clavichord creating instruments capable of producing sounds no one has ever heard.

Much as a neural network can learn to identify a cat by analyzing hundreds of cat photos, it can learn the musical characteristics of a bassoon by analyzing hundreds of notes. It creates a mathematical representation, or vector, that identifies a bassoon. So, Mr. Eck and his team have fed notes from hundreds of instruments into a neural network, building a vector for each one. Now, simply by moving a button across a screen, they can combine these vectors to create new instruments. One may be 47 percent bassoon and 53 percent clavichord. Another might switch the percentages. And so on.

For centuries, orchestral conductors have layered sounds from various instruments atop one other. But this is different. Rather than layering sounds, Mr. Eck and his team are combining them to form something that didnt exist before, creating new ways that artists can work. Were making the next film camera, Mr. Eck said. Were making the next electric guitar.

Called NSynth, this particular project is only just getting off the ground. But across the worlds of both art and technology, many are already developing an appetite for building new art through neural networks and other A.I. techniques. This work has exploded over the last few years, said Adam Ferris, a photographer and artist in Los Angeles. This is a totally new aesthetic.

In 2015, a separate team of researchers inside Google created DeepDream, a tool that uses neural networks to generate haunting, hallucinogenic imagescapes from existing photography, and this has spawned new art inside Google and out. If the tool analyzes a photo of a dog and finds a bit of fur that looks vaguely like an eyeball, it will enhance that bit of fur and then repeat the process. The result is a dog covered in swirling eyeballs.

At the same time, a number of artists like the well-known multimedia performance artist Trevor Paglen or the lesser-known Adam Ferris are exploring neural networks in other ways. In January, Mr. Paglen gave a performance in an old maritime warehouse in San Francisco that explored the ethics of computer vision through neural networks that can track the way we look and move. While members of the avant-garde Kronos Quartet played onstage, for example, neural networks analyzed their expressions in real time, guessing at their emotions.

The tools are new, but the attitude is not. Allison Parrish, a New York University professor who builds software that generates poetry, points out that artists have been using computers to generate art since the 1950s. Much like as Jackson Pollock figured out a new way to paint by just opening the paint can and splashing it on the canvas beneath him, she said, these new computational techniques create a broader palette for artists.

A year ago, David Ha was a trader with Goldman Sachs in Tokyo. During his lunch breaks he started toying with neural networks and posting the results to a blog under a pseudonym. Among other things, he built a neural network that learned to write its own Kanji, the logographic Chinese characters that are not so much written as drawn.

Soon, Mr. Eck and other Googlers spotted the blog, and now Mr. Ha is a researcher with Google Magenta. Through a project called SketchRNN, he is building neural networks that can draw. By analyzing thousands of digital sketches made by ordinary people, these neural networks can learn to make images of things like pigs, trucks, boats or yoga poses. They dont copy what people have drawn. They learn to draw on their own, to mathematically identify what a pig drawing looks like.

Then, you ask them to, say, draw a pig with a cats head, or to visually subtract a foot from a horse or sketch a truck that looks like a dog or build a boat from a few random squiggly lines. Next to NSynth or DeepDream, these may seem less like tools that artists will use to build new works. But if you play with them, you realize that they are themselves art, living works built by Mr. Ha. A.I. isnt just creating new kinds of art; its creating new kinds of artists.

Read more:

How AI Is Creating Building Blocks to Reshape Music and Art - New York Times

Musk warns the rise of AI is a greater risk than North Korea as firm he backs bests humans in online combat. Photograph: Ben Macmahon/AAP

Elon Musk has warned again about the dangers of artificial intelligence, saying that it poses vastly more risk than the apparent nuclear capabilities of North Korea does.

The Tesla and SpaceX chief executive took to Twitter to once again reiterate the need for concern around the development of AI, following the victory of Musk-led AI development over professional players of the Dota 2 online multiplayer battle game.

This is not the first time Musk has stated that AI could potentially be one of the most dangerous international developments. He said in October 2014 that he considered it humanitys biggest existential threat, a view he has repeated several times while making investments in AI startups and organisations, including OpenAI, to keep an eye on whats going on.

Musk again called for regulation, previously doing so directly to US governors at their annual national meeting in Providence, Rhode Island.

Musks tweets coincide with the testing of an AI designed by OpenAI to play the multiplayer online battle arena (Moba) game Dota 2, which successfully managed to win all its 1-v-1 games at the International Dota 2 championships against many of the worlds best players competing for a $24.8m (19m) prize fund.

The AI displayed the ability to predict where human players would deploy forces and improvise on the spot, in a game where sheer speed of operation does not correlate with victory, meaning the AI was simply better, not just faster than the best human players.

Musk backed the non-profit AI research company OpenAI in December 2015, taking up a co-chair position. OpenAIs goal is to develop AI in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return. But it is not the first group to take on human players in a gaming scenario. Googles Deepmind AI outfit, in which Musk was an early investor, beat the worlds best players in the board game Go and has its sights set on conquering the real-time strategy game StarCraft II.

Musks latest comments come after a public spat with Facebook chief executive Mark Zuckerberg over the dangers of AI, with Musk dismissing Zuckerberg as having limited understanding of the subject after the social networks head called out Musk for scaremongering over AI.

Read the rest here:

Elon Musk: AI 'vastly more risky than North Korea' - The Guardian

MITs Computer Science and Artificial Intelligence Lab (CSAIL) wants to ensure your streaming video experience stays smooth. A research team led by MIT professor Mohammad Alizadeh has developed an artificial intelligence (dubbed Pensieve) that can select the best algorithms for ensuring video streams both without interruption, and at the best possible playback quality.

The method improves upon existing tech, including the adaptive bitrate (ABR) method used by YouTube that throttles back quality to keep videos playing, albeit with pixelation and other artifacts. The AI can select different algorithms depending on what kind of network conditions a device is experiencing, cutting down on the downsides associated with any one method.

During experimentation, the CSAIL research team behind this method found that video streamed with between 10 and 30 percent less rebuffing, with 10 to 25 percent improved quality. Those gains would definitely add up to a significantly improved experience for most video viewers, especially over a long period.

The difference in CSAILs Pensieve approach vs. traditional methods is mainly in its use of a neural network instead of sticking to a strictly algorithmic-based approach. The neural net learns how to optimize through a reward system that incentivizes smoother video playback, rather than setting out defined rules about what algorithmic techniques to use when buffering video.

Researchers say the system is also potentially tweakable on the user end, depending on what they want to prioritize in playback: You could, for instance, set Pensieve to optimize for playback quality, or conversely, for playback speed, or even for conservation of data.

The team is making their project code open source for Pensieve at SIGCOMM next week in LA, and they expect that when trained on a larger data set, it could provide even greater improvements in terms of performance and quality. Theyre also now going to test applying it to VR video, since the high bitrates required for a quality experience there are well suited to the kinds of improvements Pensieve can offer.

Visit link:

MIT's AI streaming software aims to stop those video stutters - TechCrunch

In some cases, researchers are working to ensure that systems dont make mistakes on their own, as the Coast Runners boat did. Theyre also working to ensure that hackers and other bad actors cant exploit hidden holes in these systems. Researchers like Googles Ian Goodfellow, for example, are exploring ways that hackers could fool A.I. systems into seeing things that arent there.

Modern computer vision is based on what are called deep neural networks, which are pattern-recognition systems that can learn tasks by analyzing vast amounts of data. By analyzing thousands of dog photos, a neural network can learn to recognize a dog. This is how Facebook identifies faces in snapshots, and its how Google instantly searches for images inside its Photos app.

But Mr. Goodfellow and others have shown that hackers can alter images so that a neural network will believe they include things that arent really there. Just by changing a few pixels in the photo of elephant, for example, they could fool the neural network into thinking it depicts a car.

That becomes problematic when neural networks are used in security cameras. Simply by making a few marks on your face, the researchers said, you could fool a camera into believing youre someone else.

If you train an object-recognition system on a million images labeled by humans, you can still create new images where a human and the machine disagree 100 percent of the time, Mr. Goodfellow said. We need to understand that phenomenon.

Another big worry is that A.I. systems will learn to prevent humans from turning them off. If the machine is designed to chase a reward, the thinking goes, it may find that it can chase that reward only if it stays on. This oft-described threat is much further off, but researchers are already working to address it.

Mr. Hadfield-Menell and others at U.C. Berkeley recently published a paper that takes a mathematical approach to the problem. A machine will seek to preserve its off switch, they showed, if it is specifically designed to be uncertain about its reward function. This gives it an incentive to accept or even seek out human oversight.

Much of this work is still theoretical. But given the rapid progress of A.I. techniques and their growing importance across so many industries, researchers believe that starting early is the best policy.

Theres a lot of uncertainty around exactly how rapid progress in A.I. is going to be, said Shane Legg, who oversees the A.I. safety work at DeepMind. The responsible approach is to try to understand different ways in which these technologies can be misused, different ways they can fail and different ways of dealing with these issues.

An earlier version of a picture caption with this article identified the three people in the picture in the wrong order. They are Dario Amodei, standing, and from left, Paul Christiano and Geoffrey Irving.

A version of this article appears in print on August 14, 2017, on Page B1 of the New York edition with the headline: When Robots Have Minds Of Their Own.

Here is the original post:

Teaching AI Systems to Behave Themselves - New York Times