This robotic surgeon cuts cleaner than a human – mit technology review

If you’re planning to buy a Tesla, you might want to do so right about … now. That’s because, according to Bloomberg, the word on Capitol Hill is that the federal government’s forthcoming tax bill will ax an existing subsidy that has so far acted as a sweetener for anyone looking to make the switch from gas to electrons.

Currently, the Internal Revenue Service provides a tax credit of up to $7,500 for each new electric car sold, which acts as a significant incentive for buyers. The credit only applies to the first 200,000 electric vehicles sold by a manufacturer in America, though—a total which would, in the current state of things, see subsidies run out sooner for Tesla than many other manufacturers. (It’s thought to have delivered around 140,000 vehicles in the U.S. to date, which is far more than any other automaker in America.)

There’s no official word on the policy: the news comes from comments made by Representative Mike Bishop, a Michigan Republican, to reporters. But such a shift certainly aligns with President Donald Trump’s climate and energy views.

Cutting the subsidy would be a regressive step that disincentivizes the adoption of electric cars download file recovery software free. A $7,500 tax break goes a long way on an affordable electric car like a Tesla Model 3 or a Chevrolet Bolt, which have base list prices of $35,000 and $37,500, respectively. Tesla’s stock has fallen on the news, but the policy decision would hit other manufacturers harder, cementing the early lead of Elon Musk’s automaker.

The move would also come at a time when many other nations are making their own pushes to encourage the adoption of low-emissions vehicles. The U.K. has banned the sale of internal-combustion cars after 2040 and earmarked one billion pounds for subsidies to help people buy hybrid and electric cars, for instance the best free file recovery software 2017. Meanwhile, many European cities are banning diesel cars, and even China is introducing its own aggressive electric-vehicle policies.

As we’ve argued regularly and strenuously in the past, fast adoption of electric vehicles could have a dramatic impact on global emissions—but it will only be made possible through progressive policy measures. If this measure is passed as part of the new tax bill, it would send the nascent electric-vehicle industry in exactly the wrong direction.

Stanford professor Mark Jacobson has sued a prominent energy researcher and the National Academy of Sciences for defamation over a sharply-worded rebuttal of his work, shifting a heated scientific debate over renewable energy out of the journals and into the courts.

The suit, filed September 29 in a Washington, D.C., superior court, demands a retraction of a June paper in the Proceedings of the National Academy of Sciences. Jacobson seeks more than $10 million in damages from both the paper’s publisher and its lead author, Christopher Clack, who is chief executive of Vibrant Clean Energy and a former NOAA researcher.

Jacobson was the lead author of a 2015 paper in the same journal that concluded wind, solar, and hydroelectric sources alone could supply 100 percent of the U.S. grid’s needs, all at low cost.

Many other energy researchers have long argued that additional technologies, such as nuclear energy, carbon capture, and advanced storage options, will be required to decarbonize the electricity sector, particularly in a cost-competitive manner.

Earlier this year, Clack and 20 other researchers published a response arguing that, as MIT Technology Review previously reported, Jacobson’s paper "contained modeling errors and implausible assumptions that could distort public policy and spending decisions." (For more details on the researchers’ critiques, check out our earlier article on the Clack paper: " Sustainable Energy Scientists Sharply Rebut Influential Renewable-Energy Plan.")

The lawsuit claims that Clack and his peers’ rebuttal contained false and misleading information, and that its publication and subsequent press coverage harmed Jacobson’s reputation.

In an e-mailed statement, Clack said: "Our paper underwent very rigorous peer review, and two further extraordinary editorial reviews by the nation’s most prestigious academic journal, which considered Dr recover deleted files software free. Jacobson’s criticisms and found them to be without merit."

Several observers were quick to criticize Jacobson for taking the unusual step of turning to the courts to resolve a technical dispute of the sort that regularly occurs in science, but is usually haggled over in journals and at conferences.

"Using courts to resolve sci issues? Generally a bad idea," climate scientist Gavin Schmidt tweeted in response to news of Jacobson’s court filing.

Michael Shellenberger, president of pro-nuclear advocacy group Environmental Progress, wrote: "Jacobson’s lawsuit is an appalling attack on free speech and scientific inquiry and we urge the courts to reject it as grossly unethical and without legal merit."

Artificial intelligence, long used in stock picking and other financial services, is making its way into one of the most popular kind of investment instruments around.

The Globe and Mail reports that Horizons ETFs Management will launch Horizons Active AI Global Equity ETF (which stands for “exchange-traded fund”) today on the Toronto Stock Exchange with the ticker symbol MIND. This follows the debut of the very first AI-run ETF on the U.S. stock exchange last month. The AI Powered Equity ETF (ticker AIEQ) uses IBM’s Watson technology—which, let’s be honest, may not be the wisest investment strategy ever—to make picks by analyzing data from over 6,000 equities. The AI also looks at information like economic data, regulatory filings, quarterly results, headline news, and even social media.

ETFs are a popular investment fund vehicle that typically track indexes like stocks or bonds. Part of their appeal is that they can be traded like stocks and are typically cheaper to own than mutual funds or other similar products.

MIND was designed by Qraft Technologies, a South Korean financial technology firm, and will track 50 investment metrics and rebalance the investments monthly. To train the AI, Horizons gave it 10 years of investment data.

Although it only launched October 20, AIEQ has already made some bets that let it stand out from human-run funds, according to Nicholas Colas, a cofounder of DataTrek Research facebook password hack software free download rar. AIEQ has a few positions that are 2.5 to 3 percent of its portfolio each, which is larger than most managers make.

As of Wednesday morning, AIEQ is down 1.3 percent since its launch, while NASDAQ is up 0.9 percent over the same period. But it’s too early to tell if AI-managed ETFs will do better than humans. Colas points out that hedge funds each spend tens of millions of dollars hiring quants to do all kinds of analyses on mountains of financial data—which uses plenty of computing muscle, as well as sophisticated algorithms. Technology is, in other words, already a big part of stock picking, and these AI-powered ETFs are going to have their work cut out for them.

When it comes to making bucketloads of cash, any edge is worth pursuing file recovery software shareware. And for hedge funds, that now means flirting with the idea of using quantum computers in an attempt to give their analysis a speed boost.

The Financial Times reports (paywall) that hedge funds including Two Sigma, Renaissance, DE Shaw, and WorldQuant are all experimenting with quantum computing systems online data recovery software free. It also quotes the CEO of the ( controversial) quantum computing firm D-Wave as saying that the company has had “a lot” of conversations with hedge funds and banks. It’s not yet clear what kinds of devices the hedge funds are working with, though.

It’s not surprising news. Hedge funds often try out emerging technologies and approaches—from machine learning to crowdsourced coding—to find out whether they can make an extra buck (or billion). And the time is very much right to start with quantum computing: while the devices are still immature, pushes by Intel, IBM, Google, and even startups like Rigetti are slowly beginning to make them a practical reality. That’s why we named them one of our 10 Breakthrough Technologies of 2017.

What’s less clear is how useful quantum computers will actually be to hedge funds. The exotic devices are in theory well suited to rattling through some very specific optimization problems far faster than regular computers, but they may be able to find only limited application in the world of finance sandisk file recovery software free. The newspaper says that one possible application will be an optimization problem, though: “arranging in real time the best possible basket of various assets and securities for shifting market environments.”

AI has enjoyed huge growth in the past few years, and much of that success is owed to deep neural networks, which provide the smarts behind impressive tricks like image recognition. But there is growing concern that some of the fundamental principles that have made those systems so successful may not be able to overcome the major problems facing AI—perhaps the biggest of which is a need for huge quantities of data from which to learn (for a deep dive on this, check out our feature " Is AI Riding a One-Trick Pony?").

Google’s Geoff Hinton appears to be among those fretting about AI’s future. As Wired reports, Hinton has unveiled a new take on traditional neural networks that he calls capsule networks winrar file password remover software. In a pair of new papers—one published on the arXIv, the other on OpenReview—Hinton and a handful of colleagues explain how they work.

Their approach uses small groups of neurons, collectively known as capsules, which are organized into layers to identify things in video or images. When several capsules in one layer agree on having detected something, they activate a capsule at a higher level—and so on, until the network is able to make a judgment about what it sees. Each of those capsules is designed to detect a specific feature in an image in such a way that it can recognize them in different scenarios, like from varying angles.

Hinton claims that the approach, which has been in the making for decades, should enable his networks to recognize objects in new situations with less data than regular neural nets must use.

In the papers published so far, capsule networks have been shown to keep up with regular neural networks when it comes to identifying handwritten characters, and they make fewer errors when trying to recognize previously observed toys from different angles. But for the moment, at least, they’re still a bit slower than their traditional counterparts.

Now, then, comes the interesting part. Will these systems provide a compelling alternative to traditional neural networks, or will they stall? We can expect the machine-learning community to implement the work, and fast, in order to find out. Either way, those concerned about the limitations of current AI systems can be heartened by the fact that researchers are pushing the boundaries to build new deep-learning alternatives.

Researchers at Nvidia created the celeb-generating algorithm using a clever new machine-learning technique software recovery free. The faces are dreamed up using a more efficient version of what’s known as a generative adversarial network (or GAN).

A GAN consists of two neural networks, both trained using a particular data set. One network then tries to generate synthetic examples to fool the other network into thinking they came from the original data set winrar encrypted file password cracker software. The process helps the first network improve its ability to produce realistic data.

GANs were invented by Google researcher Ian Goodfellow (who is also one of our 35 Innovators Under 35 for 2017), and they have proved remarkably effective for synthesizing realistic-sounding speech and all sorts of dazzling imagery. They could prove very useful for generating animated graphics for video games, and for compressing video more efficiently.

In a paper (PDF) submitted to an upcoming conference, the Nvidia researchers claim to have developed a better GAN by having it start off working with low-resolution images, and gradually increasing the image resolution as well as the size of the networks involved. They fed their GAN a data set of celebrity faces, and it produced some very realistic-looking faces (you can check out a video of the research here).

One thing to note, however, is that a few of the images feature strange artifacts and features, like a missing eyebrow or teeth in the wrong place—not exactly things that would get you a gig on reality TV. This goes to show that even if machine learning can produce amazing visual trickery, it lacks the deeper intelligence required to make sense of the real world.