The Department of Economic and Social Affairs at the United Nations projects that 9.7 billion people will inhabit the Earth come 2050. With one in eight people today not getting enough food, farmers will have to become more prolific in order to serve these additional billions. As nearly half of the planet's 10 global hectares of potentially productive land is already devoted to agriculture, any expansion will increasingly impact delicate ecosystems that are already declining. What's worse is the World Bank estimates that climate change could cut crop yields by more than 25 percent as the population continues to grow. Feeding this amount of people will require finding new ways of becoming even more efficient at producing food.

Few places evoke the rise and fall of Japan's industrial might than the head office of the Imperial Steel Works in Kitakyushu. The red brick Meiji Era building was the heart of the nation's first big steelworks. Kitakyushu, with nearly a million people, embodies the struggle of Japan's cities to adapt to a future where citizens are older, workers are fewer and many houses are emptying. The emblems of government efforts to revitalize the economy -- a billion-dollar airport, a robotics factory -- stand beside the empty lots, an idle blast furnace and shuttered shops. Five hours west of Tokyo by shinkansen, Kitakyushu lost over 15,000 people in the five years to 2015, more than any other city in the country apart from those evacuated because of the Fukushima nuclear disaster.

Managers of US National Forests must decide what policy to apply for dealing with lightning-caused wildfires. Conflicts among stakeholders (e.g., timber companies, home owners, and wildlife biologists) have often led to spirited political debates and even violent eco-terrorism. One way to transform these conflicts into multi-stakeholder negotiations is to provide a high-fidelity simulation environment in which stakeholders can explore the space of alternative policies and understand the tradeoffs therein. Such an environment needs to support fast optimization of MDP policies so that users can adjust reward functions and analyze the resulting optimal policies. This paper assesses the suitability of SMAC---a black-box empirical function optimization algorithm---for rapid optimization of MDP policies. The paper describes five reward function components and four stakeholder constituencies. It then introduces a parameterized class of policies that can be easily understood by the stakeholders. SMAC is applied to find the optimal policy in this class for the reward functions of each of the stakeholder constituencies. The results confirm that SMAC is able to rapidly find good policies that make sense from the domain perspective. Because the full-fidelity forest fire simulator is far too expensive to support interactive optimization, SMAC is applied to a surrogate model constructed from a modest number of runs of the full-fidelity simulator. To check the quality of the SMAC-optimized policies, the policies are evaluated on the full-fidelity simulator. The results confirm that the surrogate values estimates are valid. This is the first successful optimization of wildfire management policies using a full-fidelity simulation. The same methodology should be applicable to other contentious natural resource management problems where high-fidelity simulation is extremely expensive.

Seven million years ago, the first ancestors of mankind appeared in Africa and seven million years later, as we speak, mankind's existence is being traced by archaeologists in South Africa, where they believe they are finding several missing links in our history. A history traced back to the first hominid forms. What is a hominid, I hear you say, and when did it exist? Well, way back when scientists believe that the Eurasian and American tectonic plates collided and then settled, creating a massive flat area in Africa, after the Ice Age. This new massive field was flat for hundreds of miles, as far as the eye could see, and the apes that inhabited this land suddenly found there were no trees to climb. This meant that the apes found it hard going thundering over hundreds of miles on their hands and feet, so they started to stand up to make it easier to move over the land. This resulted in a change in the wiring of the brain, which, over thousands of years, led to the early forms of what is now recognized as human. The first link to understanding this chain was the discovery of Lucy. Lucy--named after the Beatles song "Lucy in the Sky with Diamonds"--is the first skeleton that could be pieced together to show how these early human forms appeared on the African plains in the post-Ice Age world. The skeleton was found in the early 1970s in Ethiopia by paleoanthropologist Donald Johanson and is an early example of the hominid australopithecine, dating back to about 3.2 million years ago. The skeleton presents a small skull akin to that of most apes, plus evidence of a walking gait that was bipedal and upright, similar to that of humans and other hominids. This combination supports the view of human evolution that bipedalism preceded an increase in brain size. Since Lucy was found, there have been many other astonishing discoveries in what is now called the "Cradle of Humankind" in South Africa, a Unesco World Heritage site.

Eating a robot might not sound very appealing, but edible machines could be crawling their way down your intestinal tract in the near future. Scientists have created the first part of these digestible bots that could deliver medicine to people in need. The team is working with a hospitality school to see if they could even make them taste nice. The part they have created is just 90mm (3.5in) in length, 20mm (0.8in) wide and 17mm (0.7in) in thickness. This is the first part of an edible robot, which will be biodegradable and have very low levels of toxicity. Experts could train them to walk to humans who are in need of food.

March 12, 2017 --In traditional computers, the smallest units of information exists in one of two states: 1 and 0, or on and off. Long strings of 1s and 0s can store increasingly complex information that can be used use to perform useful tasks, but that information storage is limited by the size of those individual bits of information in a computer's hard drive. But now, researchers have figured out a way to magnetically store information on the smallest unit possible: a single atom. There's a long way to go before atom-sized information storage technology can make it to your home computer or smartphone, but now researchers have proven that it is possible to store information on an incredibly small level. Theoretically, this new technology could lead to massive data storage capacities on an impressive scale – even in the smallest of devices.

Context: One of the black arts of data mining is learning the magic parameters which control the learners. In software analytics, at least for defect prediction, several methods, like grid search and differential evolution (DE), have been proposed to learn these parameters, which has been proved to be able to improve the performance scores of learners. Objective: We want to evaluate which method can find better parameters in terms of performance score and runtime cost. Methods: This paper compares grid search to differential evolution, which is an evolutionary algorithm that makes extensive use of stochastic jumps around the search space. Results: We find that the seemingly complete approach of grid search does no better, and sometimes worse, than the stochastic search. When repeated 20 times to check for conclusion validity, DE was over 210 times faster than grid search to tune Random Forests on 17 testing data sets with F-Measure Conclusions: These results are puzzling: why does a quick partial search be just as effective as a much slower, and much more, extensive search? To answer that question, we turned to the theoretical optimization literature. Bergstra and Bengio conjecture that grid search is not more effective than more randomized searchers if the underlying search space is inherently low dimensional. This is significant since recent results show that defect prediction exhibits very low intrinsic dimensionality-- an observation that explains why a fast method like DE may work as well as a seemingly more thorough grid search. This suggests, as a future research direction, that it might be possible to peek at data sets before doing any optimization in order to match the optimization algorithm to the problem at hand.

Juergen Schmidhuber taught a computer to park a car. He's also showing that same machine how to trade stocks and detect flaws in steel production. Unrelated as these tasks may appear, Schmidhuber thinks a seemingly random training regimen is key to creating artificial intelligence that can solve any problem. Schmidhuber's AI theories tend to carry weight. In 1997, he co-authored a seminal paper that laid the groundwork for modern AI systems.

These days, marketers can't read about their profession without getting bombarded with wild claims about how AI is going to disrupt everything they do. And with the sales and marketing functions evolving so rapidly in recent years, marketers in particular must embrace an entrepreneurial spirit and constantly explore new technologies in order to give their team a competitive edge. That mindset shift, along with new consumer trends -- such as self-driving cars and intelligent voice-first products like Amazon's Alexa and Apple's Siri -- are bringing the possibilities of AI to the forefront of business-to-business marketing technology discussions. But all of this begs the question, "Which AI claims are hype and which are reality?" In order to know what a new technology like AI can bring to the table, it's important to fully understand the problems you're trying to solve.

Researchers from Argonne National Laboratory, using supercomputers at Berkeley Lab's National Energy Research Scientific Computing Center (NERSC), are employing machine learning algorithms to accurately predict the physical, chemical and mechanical properties of nanomaterials, reducing the time it takes to yield such predictions from years to months--in some cases even weeks. This approach could help accelerate the discovery and development of new materials. Using a modeling framework built around a molecular dynamics code (LAMMPS), the research team ran a series of simulations to study the structure and temperature-dependent thermal conductivity of stanene, a 2D material made up of a one-atom-thick sheet of tin. This work, which involved a set of parameters known as the "many-body interatomic potential" or "force field," yielded the first atomic-level computer model that accurately predicts stanene's structural, elastic and thermal properties. The findings were published in The Journal of Physical Chemistry Letters.