Abstract--Modern smart grids rely on advanced metering infrastructure (AMI) networks for monitoring and billing purposes. However, such an approach suffers from electricity theft cyberattacks. Different from the existing research that utilizes shallow, static, and customer-specific-based electricity theft detectors, this paper proposes a generalized deep recurrent neural network (RNN)-based electricity theft detector that can effectively thwart these cyberattacks. The proposed model exploits the time series nature of the customers' electricity consumption to implement a gated recurrent unit (GRU)-RNN, hence, improving the detection performance. In addition, the proposed RNN-based detector adopts a random search analysis in its learning stage to appropriately fine-tune its hyper-parameters. Extensive test studies are carried out to investigate the detector's performance using publicly available real data of 107,200 energy consumption days from 200 customers. Simulation results demonstrate the superior performance of the proposed detector compared with state-of-the-art electricity theft detectors.

We present a novel framework that enables efficient probabilistic inference in large-scale scientific models by allowing the execution of existing domain-specific simulators as probabilistic programs, resulting in highly interpretable posterior inference. Our framework is general purpose and scalable, and is based on a cross-platform probabilistic execution protocol through which an inference engine can control simulators in a language-agnostic way. We demonstrate the technique in particle physics, on a scientifically accurate simulation of the tau lepton decay, which is a key ingredient in establishing the properties of the Higgs boson. High-energy physics has a rich set of simulators based on quantum field theory and the interaction of particles in matter. We show how to use probabilistic programming to perform Bayesian inference in these existing simulator codebases directly, in particular conditioning on observable outputs from a simulated particle detector to directly produce an interpretable posterior distribution over decay pathways. Inference efficiency is achieved via inference compilation where a deep recurrent neural network is trained to parameterize proposal distributions and control the stochastic simulator in a sequential importance sampling scheme, at a fraction of the computational cost of Markov chain Monte Carlo sampling.

If the plans to allow multiple suppliers to provide single residencies become a reality, the way homes receive their power will change, and we'll see energy start to be provided in all round better and smarter ways. For example, when solar panels have fully-charged a battery in a home while the occupants are at work (and are therefore using little or no power at home), AI could facilitate the distribution and costs of supplying neighbouring buildings - eventually cutting down unused renewable power to almost zero.

Creating official, high-quality statistics based on big data and register data is not a simple matter. The data sources were not designed for statistical use and safeguarding the quality and continuity is quite complex. Therefore, the ultimate challenge for data scientists is to develop methods that'translate' huge amounts of data into high-quality statistics. CBS pursues this challenge with methods such as machine learning. As CBS data scientist Marc Ponsen explains, 'Machine learning has received an enormous boost by faster computers and the huge amounts of data that have become available, although what the best possible method is very much depends on the domain under investigation.'

The intelligent world is truly upon us, and based on the technological advancements that we are becoming used to, one cannot wait for the future to arrive. In this intelligent world, there is pressure on businesses to deliver an exceptional customer experience, and to use every technology they can to ensure that the customers get what they expect. I recently had the opportunity to attend the SAP Ariba Live event in Amsterdam. The event is one of the biggest supply chain and procurement conferences held across the globe. At the conference, I had the opportunity to hear and witness excellent use cases, and speak to the president of SAP Ariba, Barry Padgett.

Neural networks can predict where aftershocks will occur after earthquakes to a higher accuracy than standard techniques, according to a new study published in Nature on Wednesday. After a major earthquake aftershocks can endanger the survivors and those seeking to help them. Now a group of researchers from Harvard University, the University of Connecticut and Google, have tried to figure this out by training neural networks to predict if an aftershock will occur at a particular position. A simple feedforward neural network was trained using data taken from more than 131,000 pairs of main shocks and their aftershocks. The seismic wave patterns are split into grid cells measuring a fixed volume.

The opportunities that artificial intelligence (AI) can unlock for our world -- from discovering cures to diseases that kill millions each year to significantly reducing carbon emissions -- are expanding every day -- and is already enabling pathways to financial inclusion, citizen engagement, more affordable healthcare, and many more vital systems and services. The same types of machine learning systems that might have highlighted a certain post in your Facebook newsfeed based on your online activity are being leveraged, for instance, to highlight certain applicants in a hiring process. While public attention often focuses either on the existential threats artificial super-intelligence poses to humanity ("the robots are coming to kill us"), or the opposite salvation narrative (" AI will solve all our problems") there is a more immediate-but less visible- risk that our reliance on ML-driven decision making poses in terms of the reinforcement of systemic bias and discrimination. Machine learning technologies are already making life-altering decisions for human lives on a daily basis. Examples come from the New York Times: "Algorithms can decide where kids go to school… where building code inspections should be targeted, and even what metrics are used to rate a teacher."

From smartphone assistants to image recognition and translation, machine learning already helps us in our everyday lives. But it can also help us to tackle some of the world's most challenging physical problems -- such as energy consumption. Large-scale commercial and industrial systems like data centers consume a lot of energy, and while much has been done to stem the growth of energy use, there remains a lot more to do given the world's increasing need for computing power. Google is taking many steps to reduce energy consumptions . Compared to five years ago, Google now get around 3.5 times the computing power out of the same amount of energy.

With recent progress in algorithms and the availability of massive amounts of computation power, application of machine learning techniques is becoming a hot topic in the oil and gas industry. One of the most promising aspects to apply machine learning to the upstream field is the rock facies classification in reservoir characterization, which is crucial in determining the net pay thickness of reservoirs, thus a definitive factor in drilling decision making process. For complex machine learning tasks like facies classification, feature engineering is often critical. This paper shows the inclusion of physics-motivated feature interaction in feature augmentation can further improve the capability of machine learning in rock facies classification. We demonstrate this approach with the SEG 2016 machine learning contest dataset and the top winning algorithms. The improvement is roboust and can be $\sim5\%$ better than current existing best F-1 score, where F-1 is an evaluation metric used to quantify average prediction accuracy.

In this paper, we study a generic learning framework for building robust text classification model that achieves accuracy comparable to standard full models under test-time budget constraints. Our approach learns a selector to identify words that are relevant to the prediction tasks and only passes these words to the classifier for processing. The selector is trained jointly with the classifier and directly learns to incorporate with the classifier. We further propose a data aggregation scheme to improve the robustness of the classifier. Our learning framework is general and can be incorporated with any type of text classification model. On real-world data, we show that the proposed approach improves the performance of a given classifier and speeds up the model with a mere loss in accuracy performance.