Total and Google Cloud have signed an agreement to jointly develop artificial intelligence (AI) solutions applied to subsurface data analysis for oil and gas exploration and production. The agreement focuses on the development of AI programs that will make it possible to interpret subsurface images, notably from seismic studies (using Computer Vision technology) and automate the analysis of technical documents (using Natural Language Processing technology). These programs will allow Total's geologists, geophysicists, reservoir and geo-information engineers to explore and assess oil and gas fields faster and more effectively. Under this partnership, Total geoscientists will work side-by-side with Google Cloud's machine learning experts within the same project team based in Google Cloud's Advanced Solutions Lab in California. Total started applying artificial intelligence to characterize oil and gas fields using machine learning algorithms in the 1990s.

Efficient exploration is one of the key challenges for reinforcement learning (RL) algorithms. Most traditional sample efficiency bounds require strategic exploration. Recently many deep RL algorithm with simple heuristic exploration strategies that have few formal guarantees, achieve surprising success in many domains. These results pose an important question about understanding these exploration strategies such as $e$-greedy, as well as understanding what characterize the difficulty of exploration in MDPs. In this work we propose problem specific sample complexity bounds of $Q$ learning with random walk exploration that rely on several structural properties. We also link our theoretical results to some empirical benchmark domains, to illustrate if our bound gives polynomial sample complexity or not in these domains and how that is related with the empirical performance in these domains.

The paper introduces concentric Echo State Network, an approach to design reservoir topologies that tries to bridge the gap between deterministically constructed simple cycle models and deep reservoir computing approaches. We show how to modularize the reservoir into simple unidirectional and concentric cycles with pairwise bidirectional jump connections between adjacent loops. We provide a preliminary experimental assessment showing how concentric reservoirs yield to superior predictive accuracy and memory capacity with respect to single cycle reservoirs and deep reservoir models.

Coding theory is a central discipline underpinning wireline and wireless modems that are the workhorses of the information age. Progress in coding theory is largely driven by individual human ingenuity with sporadic breakthroughs over the past century. In this paper we study whether it is possible to automate the discovery of decoding algorithms via deep learning. We study a family of sequential codes parameterized by recurrent neural network (RNN) architectures. We show that creatively designed and trained RNN architectures can decode well known sequential codes such as the convolutional and turbo codes with close to optimal performance on the additive white Gaussian noise (AWGN) channel, which itself is achieved by breakthrough algorithms of our times (Viterbi and BCJR decoders, representing dynamic programing and forward-backward algorithms). We show strong generalizations, i.e., we train at a specific signal to noise ratio and block length but test at a wide range of these quantities, as well as robustness and adaptivity to deviations from the AWGN setting.

We predict both microscopic and macroscopic variables of a chaotic fluid flow using reservoir computing. In our procedure of the prediction, we assume no prior knowledge of physical model describing a fluid flow except that its behavior is complex but deterministic. We present two ways of prediction of the complex behavior; the first called partial-prediction requires continued knowledge of partial time-series data during the prediction as well as past time-series data, while the second called full-prediction requires only past time-series data as training data. For the first case, we are able to predict long-time motion of microscopic fluid variables. For the second case, we show that the reservoir dynamics constructed from only past data of energy functions can predict the future behavior of energy functions and reproduce the energy spectrum. This implies that the obtained reservoir system constructed without the knowledge of microscopic data is equivalent to the dynamical system describing macroscopic behavior of energy functions.

Ensemble weather predictions require statistical post-processing of systematic errors to obtain reliable and accurate probabilistic forecasts. Traditionally, this is accomplished with distributional regression models in which the parameters of a predictive distribution are estimated from a training period. We propose a flexible alternative based on neural networks that can incorporate nonlinear relationships between arbitrary predictor variables and forecast distribution parameters that are automatically learned in a data-driven way rather than requiring pre-specified link functions. In a case study of 2-meter temperature forecasts at surface stations in Germany, the neural network approach significantly outperforms benchmark post-processing methods while being computationally more affordable. Key components to this improvement are the use of auxiliary predictor variables and station-specific information with the help of embeddings. Furthermore, the trained neural network can be used to gain insight into the importance of meteorological variables thereby challenging the notion of neural networks as uninterpretable black boxes. Our approach can easily be extended to other statistical post-processing and forecasting problems. We anticipate that recent advances in deep learning combined with the ever-increasing amounts of model and observation data will transform the post-processing of numerical weather forecasts in the coming decade.

Graphs are ubiquitous data structures for representing interactions between entities. With an emphasis on the use of graphs to represent chemical molecules, we explore the task of learning to generate graphs that conform to a distribution observed in training data. We propose a variational autoencoder model in which both encoder and decoder are graph-structured. Our decoder assumes a sequential ordering of graph extension steps and we discuss and analyze design choices that mitigate the potential downsides of this linearization. Experiments compare our approach with a wide range of baselines on the molecule generation task and show that our method is more successful at matching the statistics of the original dataset on semantically important metrics. Furthermore, we show that by using appropriate shaping of the latent space, our model allows us to design molecules that are (locally) optimal in desired properties.

Cooling system plays a key role in modern data center. Developing an optimal control policy for data center cooling system is a challenging task. The prevailing approaches often rely on approximated system models that are built upon the knowledge of mechanical cooling, electrical and thermal management, which is difficult to design and may lead to sub-optimal or unstable performances. In this paper we propose to utilize the large amount of monitoring data in data center to optimize the control policy. To do so, we cast the cooling control policy design into an energy cost minimization problem with temperature constraints, and tab it into the emerging deep reinforcement learning (DRL) framework. Specifically, we propose an end-to-end neural control algorithm that is based on the actor-critic framework and the deep deterministic policy gradient (DDPG) technique. To improve the robustness of the control algorithm, we test various DRL related optimization techniques, such as recurrent decision making, discounted return, different neural network architectures, and different stochastic gradient descent algorithms, and adding additional constraints on the output of the policy network. We evaluate the proposed algorithms on the EnergyPlus simulation platform and on a real data trace collected from the National Super Computing Centre (NSCC) of Singapore. Our results show that the proposed end-to-end cooling control algorithm can achieve about 10% cooling cost saving on the simulation platform compared with a canonical two stage optimization algorithm; and it can achieve about 13.6% cooling energy saving on the NSCC data trace. Furthermore, it shows high accuracy in predicting the temperature of the racks (with mean absolute error 0.1 degree) and can control the temperature of the data center zone close to the predefined threshold with variation lower to 0.2 degree.

AI is the only thing that can let us see the whole world at once. Not recording it, but seeing it – creating a global real-time database of the world," says Stuart Russell, UC-Berkeley, lead of the AI for Good breakthrough team on AI and satellite imagery. The 2nd AI for Good Global Summit connected AI innovators with public and private-sector decision-makers. Four breakthrough teams – looking at satellite imagery, healthcare, smart cities, and trust in AI – set out to propose AI strategies and supporting projects to advance sustainable development. Teams were guided in this endeavour by an expert audience representing government, industry, academia and civil society.

These vertical market use cases, created by OpenFog members and contributors, showcase how fog works in industry. They were designed to provide system architects with a resource that has enough detail to plan and design a fog implementation and extract high-level requirements. Please fill out the form below for access to all of these use case studies. Subsurface imaging and monitoring in real time is crucial for understanding subsurface structures and dynamics that may pose risks or opportunities for oil, gas and geothermal exploration and production. This use case describes an architecture for integrating IoT sensor networks with fog computing and geophysical imaging technology.