The reparameterization trick enables optimizing large scale stochastic computation graphs via gradient descent. The essence of the trick is to refactor each stochastic node into a differentiable function of its parameters and a random variable with fixed distribution. After refactoring, the gradients of the loss propagated by the chain rule through the graph are low variance unbiased estimators of the gradients of the expected loss. While many continuous random variables have such reparameterizations, discrete random variables lack useful reparameterizations due to the discontinuous nature of discrete states. In this work we introduce Concrete random variables---continuous relaxations of discrete random variables. The Concrete distribution is a new family of distributions with closed form densities and a simple reparameterization. Whenever a discrete stochastic node of a computation graph can be refactored into a one-hot bit representation that is treated continuously, Concrete stochastic nodes can be used with automatic differentiation to produce low-variance biased gradients of objectives (including objectives that depend on the log-probability of latent stochastic nodes) on the corresponding discrete graph. We demonstrate the effectiveness of Concrete relaxations on density estimation and structured prediction tasks using neural networks.

Power spectral density (PSD) maps providing the distribution of RF power across space and frequency are constructed using power measurements collected by a network of low-cost sensors. By introducing linear compression and quantization to a small number of bits, sensor measurements can be communicated to the fusion center with minimal bandwidth requirements. Strengths of data- and model-driven approaches are combined to develop estimators capable of incorporating multiple forms of spectral and propagation prior information while fitting the rapid variations of shadow fading across space. To this end, novel nonparametric and semiparametric formulations are investigated. It is shown that PSD maps can be obtained using support vector machine-type solvers. In addition to batch approaches, an online algorithm attuned to real-time operation is developed. Numerical tests assess the performance of the novel algorithms.

BARCELONA, Spain--IBM has come up with a way to show off the more artistic side of its artificial intelligence capabilities while honoring one of Spain's most famous architects. For this year's Mobile World Congress in Barcelona, IBM tasked its Watson super computer with the understanding Antoni Gaudí's greatest works such as La Pedrera and la Sagrada Família to help create a unique sculpture for the annual trade show. Using its visual-recognition technology, Watson reviewed hundreds of images of Gaudí's buildings, along with additional examples of local architecture to understand trends in composition and inspirations. IBM then tapped Watson's AlchemyLanguage software to read documents about Gaudí and other architects, along with song lyrics and other documents about the city's history and culture. "What we were really trying to do at the essence was figure out if we can programmatically start to understand what the features of a particular style or architect are," said Jeff Arn, a Watson manager at IBM. Watson identified key elements of Gaudí's Catalan modernist style--known for its dream-like colors and shapes and odes to the natural world.

The highly intelligent machines that will be unleashed in the near future won't be coming for our lives. They'll be coming for our jobs. Being rendered obsolete by technology has been a concern among the flesh-and-blood set for hundreds of years -- cars put many in the horse industry out of work, for example -- but the speed and types of recent advances are about to give the issue an exceptional urgency. Previously, it was repetitive blue-collar jobs that were at risk, such as those in manufacturing. In the near future, however, the leaps in artificial intelligence will soon make it possible for machines to do all sorts of jobs, including those that require thinking skills we once believed beyond the reach of machines.

I write those words as both a means of advice, to readers, and as an admonition, to companies. Far too many businesses are not ready for this change too few are even aware that change is afoot so preparing for this event is critical to the health of the economy and the expansion of the workforce as a whole. I refer, specifically, to the rise of artificial intelligence (A.I.) and the proliferation of big data. "The power of artificial intelligence is the ability to move the role of the machine to the left in The Knowledge Funnel, away from pure manual coding by a human, to computer programs that write themselves and eventually to heuristic identification and algorithm creation at some point in the near future." In other words, consultants need to offer value that exceeds what technology will soon be able to do by itself.

More than two years after a multi-sided civil war erupted inside Yemen that allowed Al Qaeda's local franchise to amass power and seize territory, President Trump has directed the Pentagon to embark on a complicated counter-terrorism campaign. Trump's decision, just six weeks into his presidency, intends to reverse the largely unchecked expansion across southern Yemen of the group, Al Qaeda in the Arabian Peninsula. The willingness to expand counter-terrorism operations inside war-torn Yemen was the latest signal that Trump is more willing to defer to military commanders on national security policy than President Obama, who was criticized publicly by three of his four Defense secretaries and privately by uniformed officers for micromanaging the military. Over two days this week, armed drones and warplanes conducted more than 30 airstrikes against suspected Al Qaeda positions in three Yemeni provinces, marking the first U.S. attacks in the country since an ill-fated Navy SEAL raid in January that killed two dozen civilians, including women and children, Al Qaeda militants and Chief Petty Officer William "Ryan" Owens. The aerial bombardment is expected to continue into the coming week.

The latest video by robot developer Boston Dynamics, which seems to always coincide with a rise of Terminator references, has elicited a wide range of strong emotional reactions. The new robot, Handle (because it will be handling objects), brings forth a sense of awe for its speed, agility, strength, and ability to jump. At the same time, the machine's impressiveness brings forward deep-seated fears of robots-gone-wild. "This is the debut presentation of what I think will be a nightmare-inducing robot."-Boston Raibert was right in his prediction that Handle would be viewed as nightmare-inducing, with a flurry of comments online expressing a certain level of anxiety. New @BostonDynamics Handle #robot is a parkour master; I'm delighted & horrified in equal measure.

While Artificial Intelligence systems continue to improve, one AI is doing that in a unique way. Microsoft and the University of Cambridge partnered on DeepCoder, a deep learning AI meant to mimic the learning patterns found in the human brain. DeepCoder pieces together code from other programs to problem solve. Developers said it's not to mechanize human programmers but to help people create who don't have extensive coding knowledge. " Building an IPS system requires solving two problems. First, the search problem: to find consistent programs we need to search over a suitable set of possible programs. We need to define the set (i.e., the program space) and search procedure. Second, the ranking problem: if there are multiple programs consistent with the input-output examples, which one do we return? Both of these problems are dependent on the specifics of the problem formulation. Thus, the first important decision in formulating an approach to program synthesis is the choice of a Domain Specific Language."

The company in charge of the ruined Fukushima No. 1 nuclear plant revealed Thursday it needed new ideas to design robots capable of surviving the high levels of radiation inside the site's reactors, which were damaged in a 2011 earthquake and its resulting tsunami. The Tokyo Electric Power Company (Tepco) has hit a new obstacle since being tasked with cleaning up the worst nuclear incident since the 1986 Chernobyl disaster in the Soviet Union. The exploratory robot, specially designed to navigate the underwater sections of the reactor, died last month after being exposed to "unimaginable" levels of radiation nearly nine times more potent than the previous highest dose recorded. Naohiro Masuda, president of Tepco's Fukushima Daiichi Decommissioning project, told reporters that the company had to rethink its methods in order to examine and extract the hazardous material stuck in the plant's second reactor. "We should think out of the box so we can examine the bottom of the core and how melted fuel debris spread out," Masuda said, according to the Japan Times.

Economists may be underestimating the impact on labor markets of increasing automation and the rise of artificial intelligence, according to a post published on the Bank of England's staff blog on March 1. RELATED: Starsky Robotics sees'last mile' solution for driverless trucks "The potential for simultaneous and rapid disruption, coupled with the breadth of human functions that AI might replicate, may have profound implications for labor markets," BOE regional agents Mauricio Armellini and Tim Pike wrote in the Bank Underground post. "Economists should seriously consider the possibility that millions of people may be at risk of unemployment, should these technologies be widely adopted." RELATED: Robots could replace 1.7 million American truckers in the next decade Robots and intelligent machines threaten to replace workers in industries from finance to retail to haulage, with BOE Chief Economist Andrew Haldane estimating in 2015 that 15 million British jobs and 80 million in the U.S. could be lost to automation. Past periods of technological upheaval, such as the industrial revolution, may not be a useful guide as the pace of change was slower, giving society longer to mitigate the potential consequences of increasing job displacement and inequality, according to Armellini and Pike.