In this paper, we propose and study opportunistic contextual bandits - a special case of contextual bandits where the exploration cost varies under different environmental conditions, such as network load or return variation in recommendations. When the exploration cost is low, so is the actual regret of pulling a sub-optimal arm (e.g., trying a suboptimal recommendation). Therefore, intuitively, we could explore more when the exploration cost is relatively low and exploit more when the exploration cost is relatively high. Inspired by this intuition, for opportunistic contextual bandits with Linear payoffs, we propose an Adaptive Upper-Confidence-Bound algorithm (AdaLinUCB) to adaptively balance the exploration-exploitation trade-off for opportunistic learning. We prove that AdaLinUCB achieves O((log T)^2) problem-dependent regret upper bound, which has a smaller coefficient than that of the traditional LinUCB algorithm. Moreover, based on both synthetic and real-world dataset, we show that AdaLinUCB significantly outperforms other contextual bandit algorithms, under large exploration cost fluctuations.

Electronic power inverters are capable of quickly delivering reactive power to maintain customer voltages within operating tolerances and to reduce system losses in distribution grids. This paper proposes a systematic and data-driven approach to determine reactive power inverter output as a function of local measurements in a manner that obtains near optimal results. First, we use a network model and historic load and generation data and do optimal power flow to compute globally optimal reactive power injections for all controllable inverters in the network. Subsequently, we use regression to find a function for each inverter that maps its local historical data to an approximation of its optimal reactive power injection. The resulting functions then serve as decentralized controllers in the participating inverters to predict the optimal injection based on a new local measurements. The method achieves near-optimal results when performing voltage- and capacity-constrained loss minimization and voltage flattening, and allows for an efficient volt-VAR optimization (VVO) scheme in which legacy control equipment collaborates with existing inverters to facilitate safe operation of distribution networks with higher levels of distributed generation.

Stochastic bandit algorithms can be used for challenging non-convex optimization problems. Hyperparameter tuning of neural networks is particularly challenging, necessitating new approaches. To this end, we present a method that adaptively partitions the combined space of hyperparameters, context, and training resources (e.g., total number of training iterations). By adaptively partitioning the space, the algorithm is able to focus on the portions of the hyperparameter search space that are most relevant in a practical way. By including the resources in the combined space, the method tends to use fewer training resources overall. Our experiments show that this method can surpass state-of-the-art methods in tuning neural networks on benchmark datasets. In some cases, our implementations can achieve the same levels of accuracy on benchmark datasets as existing state-of-the-art approaches while saving over 50% of our computational resources (e.g. time, training iterations).

In this paper, we propose a robust sequential learning strategy for training large-scale Recommender Systems (RS) over implicit feedback mainly in the form of clicks. Our approach relies on the minimization of a pairwise ranking loss over blocks of consecutive items constituted by a sequence of non-clicked items followed by a clicked one for each user. Parameter updates are discarded if for a given user the number of sequential blocks is below or above some given thresholds estimated over the distribution of the number of blocks in the training set. This is to prevent from an abnormal number of clicks over some targeted items, mainly due to bots; or very few user interactions. Both scenarios affect the decision of RS and imply a shift over the distribution of items that are shown to the users. We provide a theoretical analysis showing that in the case where the ranking loss is convex, the deviation between the loss with respect to the sequence of weights found by the proposed algorithm and its minimum is bounded. Furthermore, experimental results on five large-scale collections demonstrate the efficiency of the proposed algorithm with respect to the state-of-the-art approaches, both regarding different ranking measures and computation time.

Earlier in my career, I spent more than a decade working in the infrastructure assurance field, more commonly referred to as critical infrastructure protection (CIP) in the years following the September 11 attacks. The goal of CIP was to identify dependencies on specific infrastructure assets, based on operational requirements. In short: "What do we need to do our job?" While the term "critical infrastructure" is often used in a generic sense, criticality is a term of art central to successful infrastructure assurance. In order to allocate resources and develop effective risk management and mitigation plans, it was necessary to identify critical assets based on operational need by answering the basic question: "What do you need to do, when do you need to do it, and where do you need to do it?"

Ice is the enemy of turbines everywhere. Some wind farms report energy production losses of up to 20 percent due to icing, according to Canadian wind-industry consultancy firm TechnoCentre Éolien (TCE), and that's not the worst of it. Over time, ice shedding from blades can damage other blades or overstress internal components, necessitating costly repairs. There's a clear and present use case, then, for an AI system that detects wind turbine icing. Fortunately, that's just what a team of researchers recently described in a paper published on the preprint server Arxiv.org

Journal of Petroleum Exploration and Production Technology manuscript No. (will be inserted by the editor) Abstract In this paper we present a data-driven model for forecasting the production increase after hydraulic fracturing (HF). We use data from fracturing jobs performed at one of the Siberian oilfields. The data includes features, characterizing the jobs, and a geological information. To predict an oil rate after the fracturing machine learning (ML) technique was applied. The MLbased prediction is compared to a prediction based on the experience of reservoir and production engineers responsible for the HFjob planning.

We propose a novel appearance-based gesture recognition algorithm using compressed domain signal processing techniques. Gesture features are extracted directly from the compressed measurements, which are the block averages and the coded linear combinations of the image sensor's pixel values. We also improve both the computational efficiency and the memory requirement of the previous DTW-based K-NN gesture classifiers. Both simulation testing and hardware implementation strongly support the proposed algorithm.

Traditional load analysis is facing challenges with the new electricity usage patterns due to demand response as well as increasing deployment of distributed generations, including photovoltaics (PV), electric vehicles (EV), and energy storage systems (ESS). At the transmission system, despite of irregular load behaviors at different areas, highly aggregated load shapes still share similar characteristics. Load clustering is to discover such intrinsic patterns and provide useful information to other load applications, such as load forecasting and load modeling. This paper proposes an efficient submodular load clustering method for transmission-level load areas. Robust principal component analysis (R-PCA) firstly decomposes the annual load profiles into low-rank components and sparse components to extract key features. A novel submodular cluster center selection technique is then applied to determine the optimal cluster centers through constructed similarity graph. Following the selection results, load areas are efficiently assigned to different clusters for further load analysis and applications. Numerical results obtained from PJM load demonstrate the effectiveness of the proposed approach.

Humanity uses a lot of energy, and while solar power here on Earth is doing a reasonable job of contributing to the power mix, scientists have long hypothesized that solar power gathered from space itself would be an altogether more effective scenario. And now China says it's going to be the first to do exactly that, announcing plans to build a solar power station that will orbit the Earth at 36,000 kilometers. According to China's state-backed Science and Technology Daily, Chinese scientists plan to build and launch small power stations into the stratosphere between 2021 and 2025, with a megawatt-level station planned for 2030 and a gigawatt-level facility before 2050. Without interference from the atmosphere or seasonal and night time loss of sunlight, these space-based solar farms could provide an inexhaustible source of clean energy, with the China Academy of Space Technology Corporation claiming such a set-up could reliably supply 99 percent of time at six times the intensity of solar plants on Earth. There are, of course, numerous challenges associated with this sci-fi-sounding plan.