This blog post was authored by Peter Cooper, Senior Product Manager, Microsoft IoT. From smart factories and smart cities to virtual personal assistants and self-driving cars, artificial intelligence (AI) and the Internet of Things (IoT) are transforming how people around the world live, work, and play. But fundamentally changing the ways people, devices, and data interact is not simple or easy work. Microsoft's AI & IoT Insider Labs was created to help all types of organizations accelerate their digital transformation. Member organizations around the world get access to support both technology development and product commercialization, for everything from hardware design to manufacturing to building applications and turning data into insights using machine learning.

Has machine learning and artificial intelligence (AI) reached to the point where they can power businesses to make better decisions and improve operational efficiencies? This is one of the common questions concerning business owners today. Machine learning systems are developing at a faster pace as they can now be trained using vast pools of available data. One powerful example is Google's demonstration of its AI-powered Google Assistant making restaurant reservations through a phone call. Marketers in the US have already started incorporating AI and seeing benefits such as improved sales and ROI (return on investment) in advertising and innovation, according to a survey conducted by LoopMe.

With the rapid growth of digitization, it can easily be argued that data has become the most valuable asset in the world. Organizations are steadily moving to insight-driven models, with business decisions, process enhancements, and technology investments all being driven by insights gained from data. Enormous budgets are being spent on trying to make sense of the abundant data available and this is only set to increase. According to a recent IDC report, it is estimated that by 2025 the Global Datasphere will grow to 175 zettabytes (175 trillion gigabytes). It also states that 60% of this data will be created and managed by businesses, driven by the growing reach of Artificial Intelligence (AI), the Internet of Things (IoT), and Machine Learning (ML), among others.

This paper presents a widely applicable approach to solving (multi-marginal, martingale) optimal transport and related problems via neural networks. The core idea is to penalize the optimization problem in its dual formulation and reduce it to a finite dimensional one which corresponds to optimizing a neural network with smooth objective function. We present numerical examples from optimal transport, martingale optimal transport, portfolio optimization under uncertainty and generative adversarial networks that showcase the generality and effectiveness of the approach.

The estimation of unknown values of parameters (or hidden variables, control variables) that characterise a physical system often relies on the comparison of measured data with synthetic data produced by some numerical simulator of the system as the parameter values are varied. This process often encounters two major difficulties: the generation of synthetic data for each considered set of parameter values can be computationally expensive if the system model is complicated; and the exploration of the parameter space can be inefficient and/or incomplete, a typical example being when the exploration becomes trapped in a local optimum of the objection function that characterises the mismatch between the measured and synthetic data. A method to address both these issues is presented, whereby: a surrogate model (or proxy), which emulates the computationally expensive system simulator, is constructed using deep recurrent networks (DRN); and a nested sampling (NS) algorithm is employed to perform efficient and robust exploration of the parameter space. The analysis is performed in a Bayesian context, in which the samples characterise the full joint posterior distribution of the parameters, from which parameter estimates and uncertainties are easily derived. The proposed approach is compared with conventional methods in some numerical examples, for which the results demonstrate that one can accelerate the parameter estimation process by at least an order of magnitude.

Stacking is a general approach for combining multiple models toward greater predictive accuracy. It has found various application across different domains, ensuing from its meta-learning nature. Our understanding, nevertheless, on how and why stacking works remains intuitive and lacking in theoretical insight. In this paper, we use the stability of learning algorithms as an elemental analysis framework suitable for addressing the issue. To this end, we analyze the hypothesis stability of stacking, bag-stacking, and dag-stacking and establish a connection between bag-stacking and weighted bagging. We show that the hypothesis stability of stacking is a product of the hypothesis stability of each of the base models and the combiner. Moreover, in bag-stacking and dag-stacking, the hypothesis stability depends on the sampling strategy used to generate the training set replicates. Our findings suggest that 1) subsampling and bootstrap sampling improve the stability of stacking, and 2) stacking improves the stability of both subbagging and bagging.

Advances in weather forecasting are helping to improve environmental forecast, for example, of wildfire activity. Weather forecasting provides numerous societal benefits, from extreme weather warnings to agricultural planning. In recent decades, advances in forecasting have been rapid, arising from improved observations and models, and better integration of these through data assimilation and related techniques. Further improvements are not yet constrained by limits on predictability. Better forecasting, in turn, can contribute to a wide range of environmental forecasting, from forest-fire smoke to bird migrations.

LOS ALAMOS, N.M., Dec. 17, 2018--Machine-learning research published in two related papers today in Nature Geosciences reports the detection of seismic signals accurately predicting the Cascadia fault's slow slippage, a type of failure observed to precede large earthquakes in other subduction zones. Los Alamos National Laboratory researchers applied machine learning to analyze Cascadia data and discovered the megathrust broadcasts a constant tremor, a fingerprint of the fault's displacement. More importantly, they found a direct parallel between the loudness of the fault's acoustic signal and its physical changes. Cascadia's groans, previously discounted as meaningless noise, foretold its fragility. "Cascadia's behavior was buried in the data. Until machine learning revealed precise patterns, we all discarded the continuous signal as noise, but it was full of rich information. We discovered a highly predictable sound pattern that indicates slippage and fault failure," said Los Alamos scientist Paul Johnson. "We also found a precise link between the fragility of the fault and the signal's strength, which can help us more accurately predict a megaquake."

All this fuss is not without good reason. They make oxygen for breathing, suck up CO₂, provide shade, reduce noise pollution, and just look at them -- they're beautiful! The thing is, though, that trees are pretty hard to map. The 124,795 trees in the San Francisco Urban Forest Map shown below, for example, were cataloged over a year of survey work by a team of certified arborists. The database they created is thorough, with information on tree species and size as well as environmental factors like the presence of power lines or broken pavement.

The utilities industry is a sector that relies on providing consistent access to things like electricity and water. Otherwise, households and enterprise-level customers could discover that they can't access the resources that should help them live and do business. Together, artificial intelligence (AI) and data science are causing positive developments for the utilities providers that choose to investigate these things. Here are some examples of technology at work. People increasingly feel excited about the potential for renewable energy. In addition to sustainability benefits, some individuals want to take advantage of federal or state tax credits for products such as solar power equipment.