Deep learning (DL), a new-generation artificial neural network research, has made profound strides in recent years. This review paper is intended to provide water resources scientists with a simple technical overview, trans-disciplinary progress update, and potentially inspirations about DL. Effective architectures, more accessible data, advances in regularization, and new computing power enabled the success of DL. A trans-disciplinary review reveals that DL is rapidly transforming myriad scientific disciplines including high-energy physics, astronomy, chemistry, genomics and remote sensing, where systematic DL toolkits, innovative customizations, and sub-disciplines have emerged. However, with a few exceptions, its adoption in hydrology has so far been gradual. The literature suggests that novel regularization techniques can effectively prevent high-capacity deep networks from overfitting. As a result, in most scientific disciplines, DL models demonstrated superior predictive and generalization performance to conventional methods. Meanwhile, less noticed is that DL may also serve as a scientific exploratory tool. A new area termed "AI neuroscience", has been born. This budding sub-discipline is accumulating a significant body of work, e.g., distilling knowledge obtained in DL networks to interpretable models, attributing decisions to inputs via back-propagation of relevance, or visualization of activations. These methods are designed to interpret the decision process of deep networks and derive insights. While scientists so far have mostly been using customized, ad-hoc methods for interpretation, vast opportunities await for DL to propel advancement in water science.

IBM has unveiled next-generation Power Systems Servers incorporating its newly designed Power9 processor, built specifically for compute-intensive AI workloads. Tthe new Power9 systems are capable of improving the training times of deep learning frameworks by nearly 4-times, allowing enterprises to build more accurate AI applications, faster. The new Power9 -based AC922 Power Systems are the first to embed PCI-Express 4.0, next-generation NVIDIA NVLink and OpenCAPI, which combined can accelerate data The system was designed to drive demonstrable performance improvements across popular AI frameworks such as Chainer, TensorFlow and Caffe, as well as accelerated databases such as Kinetica. As a result, data scientists can build applications faster, ranging from deep learning insights in scientific research, real-time fraud detection and credit risk analysis. Power9 is at the heart of the soon-to-be most powerful data-intensive supercomputers in the world, the US Department of Energy's "Summit" and "Sierra" supercomputers, and has been tapped by Google.

Analysts and journalists describe the radical economic and cultural shifts that come with digital transformation as a result of the rapid changes resulting from the adoption of new technologies. The sharing economy, refrigerators connected to the internet and self-driving cars are indeed very interesting. But less obvious is the potential for digital transformation in the industrial sector. The tier-two automotive components vendor, the oil and gas drilling contractor, the company manufacturing machine tools -- all of these companies can benefit significantly from technologies that are part of the industrial internet of things (IIoT). Industrial companies have been on the forefront of IIoT for decades, collecting data from connected industrial equipment and using it to inform maintenance activities and industrial automation systems.

Count-based exploration algorithms are known to perform near-optimally when used in conjunction with tabular reinforcement learning (RL) methods for solving small discrete Markov decision processes (MDPs). It is generally thought that count-based methods cannot be applied in high-dimensional state spaces, since most states will only occur once. Recent deep RL exploration strategies are able to deal with high-dimensional continuous state spaces through complex heuristics, often relying on optimism in the face of uncertainty or intrinsic motivation. In this work, we describe a surprising finding: a simple generalization of the classic count-based approach can reach near state-of-the-art performance on various high-dimensional and/or continuous deep RL benchmarks. States are mapped to hash codes, which allows to count their occurrences with a hash table. These counts are then used to compute a reward bonus according to the classic count-based exploration theory. We find that simple hash functions can achieve surprisingly good results on many challenging tasks. Furthermore, we show that a domain-dependent learned hash code may further improve these results. Detailed analysis reveals important aspects of a good hash function: 1) having appropriate granularity and 2) encoding information relevant to solving the MDP. This exploration strategy achieves near state-of-the-art performance on both continuous control tasks and Atari 2600 games, hence providing a simple yet powerful baseline for solving MDPs that require considerable exploration.

The concept of artificial intelligence (AI) involves machines learning and acting on data sets without human programming or intervention. AI can be broken down into machine learning, deep learning and neural networks. Without getting too technical, essentially the whole premise of AI is a machine mimicking the human brain. The machine can learn and adapt to different scenarios, and as times passes the machine, gets smarter and reacts differently to achieve better results. AI will play a pivotal role in many industries, through business intelligence and solving problems quicker than humans.

In retail, there are predictable yet dramatic time-dependent patterns in customer behavior, such as periodic changes in the number of visitors, or increases in visitors just before major holidays. The current paradigm of multi-armed bandit analysis does not take these known patterns into account. This means that for applications in retail, where prices are fixed for periods of time, current bandit algorithms will not suffice. This work provides a remedy that takes the time-dependent patterns into account, and we show how this remedy is implemented in the UCB and {\epsilon}-greedy methods and we introduce a new policy called the variable arm pool method. In the corrected methods, exploitation (greed) is regulated over time, so that more exploitation occurs during higher reward periods, and more exploration occurs in periods of low reward. In order to understand why regret is reduced with the corrected methods, we present a set of bounds that provide insight into why we would want to exploit during periods of high reward, and discuss the impact on regret. Our proposed methods perform well in experiments, and were inspired by a high-scoring entry in the Exploration and Exploitation 3 contest using data from Yahoo! Front Page. That entry heavily used time-series methods to regulate greed over time, which was substantially more effective than other contextual bandit methods.

Stochastic matrix factorization (SMF) can be regarded as a restriction of non-negative matrix factorization (NMF). SMF is useful for inference of topic models, NMF for binary matrices data, Markov chains, and Bayesian networks. However, SMF needs strong assumptions to reach a unique factorization and its theoretical prediction accuracy has not yet been clarified. In this paper, we study the maximum the pole of zeta function (real log canonical threshold) of a general SMF and derive an upper bound of the generalization error in Bayesian inference. The results give a foundation for a widely applicable and rigorous factorization method of SMF and mean that the generalization error in SMF becomes smaller than regular statistical models by Bayesian inference.

Tesla and SpaceX CEO Elon Musk is renowned for making dire predictions about how artificial intelligence will be a threat to humankind. While it's not yet self-evolving to the point of being an imminent danger, in 2017 AI did grow like crazy. At least that's the topic several CEOs wanted to mention when asked what they saw as the biggest trends in tech this year. "Automation and artificial intelligence (AI) have a symbiotic relationship and pose both a blessing and a threat to the future of commerce. With innovations in data mining and automation, the ability for large companies to mine data is vastly improving. This is creating a shift from helpful automation towards immersed AI and immense amounts of data are fueling predictive consumerism that furthers the ability for large corporations to accurately predict and recommend products, shaping consumer habits. This paves the way for traditional verticals of commerce and manufacturing to dissolve, and large data aggregators move directly into manufacturing. "Social listening provides brands with a bird's eye view of trending topics, sentiment and conversations related to their industry, products and public perception.

In an era marked by rapid advances in automation and artificial intelligence, new research assesses the jobs lost and jobs gained under different scenarios through 2030. The technology-driven world in which we live is a world filled with promise but also challenges. Cars that drive themselves, machines that read X-rays, and algorithms that respond to customer-service inquiries are all manifestations of powerful new forms of automation. Yet even as these technologies increase productivity and improve our lives, their use will substitute for some work activities humans currently perform--a development that has sparked much public concern. Building on our January 2017 report on automation, McKinsey Global Institute's latest report, Jobs lost, jobs gained: Workforce transitions in a time of automation (PDF–5MB), assesses the number and types of jobs that might be created under different scenarios through 2030 and compares that to the jobs that could be lost to automation. The results reveal a rich mosaic of potential shifts in occupations in the years ahead, with important implications for workforce skills and wages. Our key finding is that while there may be enough work to maintain full employment to 2030 under most scenarios, the transitions will be very challenging--matching or even exceeding the scale of shifts out of agriculture and manufacturing we have seen in the past.

Current domain-independent, classical planners require symbolic models of the problem domain and instance as input, resulting in a knowledge acquisition bottleneck. Meanwhile, although deep learning has achieved significant success in many fields, the knowledge is encoded in a subsymbolic representation which is incompatible with symbolic systems such as planners. We propose LatPlan, an unsupervised architecture combining deep learning and classical planning. Given only an unlabeled set of image pairs showing a subset of transitions allowed in the environment (training inputs), and a pair of images representing the initial and the goal states (planning inputs), LatPlan finds a plan to the goal state in a symbolic latent space and returns a visualized plan execution. The contribution of this paper is twofold: (1) State Autoencoder, which finds a propositional state representation of the environment using a Variational Autoencoder. It generates a discrete latent vector from the images, based on which a PDDL model can be constructed and then solved by an off-the-shelf planner. (2) Action Autoencoder / Discriminator, a neural architecture which jointly finds the action symbols and the implicit action models (preconditions/effects), and provides a successor function for the implicit graph search. We evaluate LatPlan using image-based versions of 3 planning domains: 8-puzzle, Towers of Hanoi and LightsOut.