Simultaneous machine translation (SiMT) generates translation before reading the entire source sentence and hence it has to trade off between translation quality and latency. To fulfill the requirements of different translation quality and latency in practical applications, the previous methods usually need to train multiple SiMT models for different latency levels, resulting in large computational costs. In this paper, we propose a universal SiMT model with Mixture-of-Experts Wait-k Policy to achieve the best translation quality under arbitrary latency with only one trained model. Specifically, our method employs multi-head attention to accomplish the mixture of experts where each head is treated as a wait-k expert with its own waiting words number, and given a test latency and source inputs, the weights of the experts are accordingly adjusted to produce the best translation. Experiments on three datasets show that our method outperforms all the strong baselines under different latency, including the state-of-the-art adaptive policy.

Are you ready to flex your Deep Learning skills by learning how to build and implement an Artificial Neural Network using Python from scratch? Testing your skills with practical courses is one of the best and most enjoyable ways to learn data science…and now we're giving you that chance for FREE. In this free course, AI expert Hadelin de Ponteves guides you through a case study that shows you how to build an ANN Regression model to predict the electrical energy output of a Combined Cycle Power Plant. The objective is to create a data model that predicts the net hourly electrical energy output (EP) of the plant using available hourly average ambient variables. Go hands-on with Hadelin in solving this complex, real-world Deep Learning challenge that covers everything from data preprocessing to building and training an ANN, while utilizing the Machine Learning library, Tensorflow 2.0, and Google Colab, the free, browser-based notebook environment that runs completely in the cloud.

The progress of some AI paradigms such as deep learning is said to be linked to an exponential growth in the number of parameters. There are many studies corroborating these trends, but does this translate into an exponential increase in energy consumption? In order to answer this question we focus on inference costs rather than training costs, as the former account for most of the computing effort, solely because of the multiplicative factors. Also, apart from algorithmic innovations, we account for more specific and powerful hardware (leading to higher FLOPS) that is usually accompanied with important energy efficiency optimisations. We also move the focus from the first implementation of a breakthrough paper towards the consolidated version of the techniques one or two year later. Under this distinctive and comprehensive perspective, we study relevant models in the areas of computer vision and natural language processing: for a sustained increase in performance we see a much softer growth in energy consumption than previously anticipated. The only caveat is, yet again, the multiplicative factor, as future AI increases penetration and becomes more pervasive.

Within the national innovation system literature, empirical analyses are severely lacking for developing economies. Particularly, the low- and middle-income countries (LMICs) eligible for the World Bank's International Development Association (IDA) support, are rarely part of any empirical discourse on growth, development, and innovation. One major issue hindering panel analyses in LMICs, and thus them being subject to any empirical discussion, is the lack of complete data availability. This work offers a new complete panel dataset with no missing values for LMICs eligible for IDA's support. I use a standard, widely respected multiple imputation technique (specifically, Predictive Mean Matching) developed by Rubin (1987). This technique respects the structure of multivariate continuous panel data at the country level. I employ this technique to create a large dataset consisting of many variables drawn from publicly available established sources. These variables, in turn, capture six crucial country-level capacities: technological capacity, financial capacity, human capital capacity, infrastructural capacity, public policy capacity, and social capacity. Such capacities are part and parcel of the National Absorptive Capacity Systems (NACS). The dataset (MSK dataset) thus produced contains data on 47 variables for 82 LMICs between 2005 and 2019. The dataset has passed a quality and reliability check and can thus be used for comparative analyses of national absorptive capacities and development, transition, and convergence analyses among LMICs.

Ultrasonic methods have great potential applications to detect and characterize defects in multi-layered bonded composites. However, it remains challenging to quantitatively reconstruct defects, such as disbonds and kissing bonds, that influence the integrity of adhesive bonds and seriously reduce the strength of assemblies. In this work, an ultrasonic method based on the supervised fully convolutional network (FCN) is proposed to quantitatively reconstruct defects hidden in multi-layered bonded composites. In the training process of this method, an FCN establishes a non-linear mapping from measured ultrasonic data to the corresponding velocity models of multi-layered bonded composites. In the predicting process, the trained network obtained from the training process is used to directly reconstruct the velocity models from the new measured ultrasonic data of adhesively bonded composites. The presented FCN-based inversion method can automatically extract useful features in multi-layered composites. Although this method is computationally expensive in the training process, the prediction itself in the online phase takes only seconds. The numerical results show that the FCN-based ultrasonic inversion method is capable to accurately reconstruct ultrasonic velocity models of the high contrast defects, which has great potential for online detection of adhesively bonded composites.

Digital technologies and artificial intelligence (AI) in particular have tremendous potential to deliver the energy sector's climate goals more rapidly and at lower cost, according to a new study. As the impacts of climate change become more visible worldwide, governments and industry face the urgent challenge of transitioning to a low-carbon global energy system. The World Economic Forum's report highlights the technologies' potential to support the energy transition and advances a set of principles for the energy industry to deploy AI in a "safe, fair, and trustworthy way". Produced in collaboration with BloombergNEF (BNEF) and German energy agency Deutsche Energie-Agentur (Dena), Harnessing AI to accelerate the energy transition reviews the state of play of AI adoption in the energy sector. It also identifies high-priority applications of AI in the energy transition and offers a road map and practical recommendations for the energy and AI industries to maximise AI's benefits.

Virtual flow metering (VFM) is a cost-effective and non-intrusive technology for inferring multiphase flow rates in petroleum assets. Inferences about flow rates are fundamental to decision support systems that operators extensively rely on. Data-driven VFM, where mechanistic models are replaced with machine learning models, has recently gained attention due to its promise of lower maintenance costs. While excellent performances in small sample studies have been reported in the literature, there is still considerable doubt about the robustness of data-driven VFM. In this paper, we propose a new multi-task learning (MTL) architecture for data-driven VFM. Our method differs from previous methods in that it enables learning across oil and gas wells. We study the method by modeling 55 wells from four petroleum assets and compare the results with two single-task baseline models. Our findings show that MTL improves robustness over single-task methods, without sacrificing performance. MTL yields a 25-50% error reduction on average for the assets where single-task architectures are struggling.

A lowering in the cost of batteries and solar PV systems has led to a high uptake of solar battery home systems. In this work, we use the deep deterministic policy gradient algorithm to optimise the charging and discharging behaviour of a battery within such a system. Our approach outputs a continuous action space when it charges and discharges the battery, and can function well in a stochastic environment. We show good performance of this algorithm by lowering the expenditure of a single household on electricity to almost \$1AUD for large batteries across selected weeks within a year.

The transition to a renewable energy system poses challenges for power grid operation and stability. Secondary control is key in restoring the power system to its reference following a disturbance. Underestimating the necessary control capacity may require emergency measures, such as load shedding. Hence, a solid understanding of the emerging risks and the driving factors of control is needed. In this contribution, we establish an explainable machine learning model for the activation of secondary control power in Germany. Training gradient boosted trees, we obtain an accurate description of control activation. Using SHapely Additive exPlanation (SHAP) values, we investigate the dependency between control activation and external features such as the generation mix, forecasting errors, and electricity market data. Thereby, our analysis reveals drivers that lead to high reserve requirements in the German power system. Our transparent approach, utilizing open data and making machine learning models interpretable, opens new scientific discovery avenues.

The rational design of molecules with desired properties is a long-standing challenge in chemistry. Generative neural networks have emerged as a powerful approach to sample novel molecules from a learned distribution. Here, we propose a conditional generative neural network for 3d molecular structures with specified structural and chemical properties. This approach is agnostic to chemical bonding and enables targeted sampling of novel molecules from conditional distributions, even in domains where reference calculations are sparse. We demonstrate the utility of our method for inverse design by generating molecules with specified composition or motifs, discovering particularly stable molecules, and jointly targeting multiple electronic properties beyond the training regime.