When we think about technology, we often think about physical devices that are electrical or digital. In fact technology encompasses far more than that. The dictionary definition refers to Technology as, "methods, systems, and devices which are the result of scientific knowledge being used for practical purposes." As we look to the year ahead tech disruption will be driven as much by the methods and systems as it is by the devices we associate with tech disruption. It's impossible to predict exactly which trends will become the most disruptive over the course of 2018.

Mathematicians at the Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab) have developed a new approach to machine learning aimed at experimental imaging data. Rather than relying on the tens or hundreds of thousands of images used by typical machine learning methods, this new approach "learns" much more quickly and requires far fewer images. Daniël Pelt and James Sethian of Berkeley Lab's Center for Advanced Mathematics for Energy Research Applications (CAMERA) turned the usual machine learning perspective on its head by developing what they call a "Mixed-Scale Dense Convolution Neural Network (MS-D)" that requires far fewer parameters than traditional methods, converges quickly, and has the ability to "learn" from a remarkably small training set. Their approach is already being used to extract biological structure from cell images, and is poised to provide a major new computational tool to analyze data across a wide range of research areas. As experimental facilities generate higher resolution images at higher speeds, scientists can struggle to manage and analyze the resulting data, which is often done painstakingly by hand.

Recent advances in deep reinforcement learning have made significant strides in performance on applications such as Go and Atari games. However, developing practical methods to balance exploration and exploitation in complex domains remains largely unsolved. Thompson Sampling and its extension to reinforcement learning provide an elegant approach to exploration that only requires access to posterior samples of the model. At the same time, advances in approximate Bayesian methods have made posterior approximation for flexible neural network models practical. Thus, it is attractive to consider approximate Bayesian neural networks in a Thompson Sampling framework. To understand the impact of using an approximate posterior on Thompson Sampling, we benchmark well-established and recently developed methods for approximate posterior sampling combined with Thompson Sampling over a series of contextual bandit problems. We found that many approaches that have been successful in the supervised learning setting underperformed in the sequential decision-making scenario. In particular, we highlight the challenge of adapting slowly converging uncertainty estimates to the online setting.

The MIT Technology Review has released a list of technologies it believes will make the most impact over the next 12 months, including smarter cities, genetic fortune telling and "babel fish" earphones.

The future of work is going to be determined by artificial intelligence and automation. These technologies will eliminate some jobs, but they will also create new opportunities and greater demand for the jobs that humans still do best. We decided to shine the spotlight on five positions you will see much more of on job boards in 2018. Want to keep up with job growth, automation and the future of work? Sign up for our new newsletter, Clocking In, which launches this month.

Mathematicians at the Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab) have developed a new approach to machine learning aimed at experimental imaging data. Rather than relying on the tens or hundreds of thousands of images used by typical machine learning methods, this new approach "learns" much more quickly and requires far fewer images. Daniël Pelt and James Sethian of Berkeley Lab's Center for Advanced Mathematics for Energy Research Applications (CAMERA) turned the usual machine learning perspective on its head by developing what they call a "Mixed-Scale Dense Convolution Neural Network (MS-D)" that requires far fewer parameters than traditional methods, converges quickly, and has the ability to "learn" from a remarkably small training set. Their approach is already being used to extract biological structure from cell images, and is poised to provide a major new computational tool to analyze data across a wide range of research areas. As experimental facilities generate higher resolution images at higher speeds, scientists can struggle to manage and analyze the resulting data, which is often done painstakingly by hand.

We've been hearing for years about the digital transformation of multiple business sectors. And the technology sector is hardly alone in this regard: For the past decade, legacy industries have been making the transition, too,and the result has been massive opportunities for entrepreneurs.

The state of Oklahoma has witnessed a stunning rise in the frequency of earthquakes, which has been linked to an increase in the use of fracking technology in the oil and gas sector. Starting in 2009, the annual number of quakes measuring above magnitude 3.0 in the state exploded from fewer than three to as many as 903 in 2015.

Computers can beat chess champions, simulate star explosions, and forecast global climate. We are even teaching them to be infallible problem-solvers and fast learners. And now, physicists at the Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab) and their collaborators have demonstrated that computers are ready to tackle the universe's greatest mysteries. The team fed thousands of images from simulated high-energy particle collisions to train computer networks to identify important features. The researchers programmed powerful arrays known as neural networks to serve as a sort of hivelike digital brain in analyzing and interpreting the images of the simulated particle debris left over from the collisions.

Every day a new blockchain-based ICO investment opportunities are emerging on the market, which is fantastic (as everyone could find something for her/his own taste)) and on the other hand, the question arises as to whether this is just about investing in that ICO and as soon as it starts to trade (appears on a virtual stock exchange) we immediately flip it (we sell bigger prices than we've bought)? The "flipping" itself should have its own a separate post entry because it is directly contributing to the drop of the value of a token issued by our honored and loved ICO. Of course, our selfish human qualities are responsible for everything. But what if we had a smarter artificial intelligence without the human selfishness? What if an Artificial Intelligence whose main objective was to improve human society would have the authority to make investment decisions?