Companies have a fast growing amounts of data to process and store, a data explosion is happening next to us. Currently one of the most common approaches to treat these vast data quantities are based on the MapReduce parallel programming paradigm. While its use is widespread in the industry, ensuring performance constraints, while at the same time minimizing costs, still provides considerable challenges. We propose a coarse grained control theoretical approach, based on techniques that have already proved their usefulness in the control community. We introduce the first algorithm to create dynamic models for Big Data MapReduce systems, running a concurrent workload. Furthermore, we identify two important control use cases: relaxed performance-minimal resource and strict performance. For the first case we develop two feedback control mechanism. A classical feedback controller and an even-based feedback, that minimises the number of cluster reconfigurations as well. Moreover, to address strict performance requirements a feedforward predictive controller that efficiently suppresses the effects of large workload size variations is developed. All the controllers are validated online in a benchmark running in a real 60 node MapReduce cluster, using a data intensive Business Intelligence workload. Our experiments demonstrate the success of the control strategies employed in assuring service time constraints.

Classification algorithms have been widely adopted to detect anomalies for various systems, e.g., IoT, cloud and face recognition, under the common assumption that the data source is clean. However, data collected from the wild can be unreliable due to careless annotations or malicious data transformation for incorrect anomaly detection. In this paper, we extend a two-layer on-line data selection framework: Robust Anomaly Detector (RAD) with a newly designed ensemble prediction where both layers contribute to the final anomaly detection decision. To adapt to the on-line nature of anomaly detection, we consider additional features of conflicting opinions of classifiers, repetitive cleaning, and oracle knowledge. We on-line learn from the incoming data streams and continuously cleanse the data, so as to adapt to the increasing learning capacity from the larger accumulated data set. Moreover, we explore the concept of oracle learning that provides additional information of true labels for difficult data points. The proposed RAD and its extensions are general and can be applied to different anomaly detection algorithms.

Convolutional Neural Network (CNN) has become the most used method for image classification tasks. During its training the learning rate and the gradient are two key factors to tune for influencing the convergence speed of the model. Usual learning rate strategies are time-based i.e. monotonous decay over time. Recent state-of-the-art techniques focus on adaptive gradient algorithms i.e. Adam and its versions. In this paper we consider an online learning scenario and we propose two Event-Based control loops to adjust the learning rate of a classical algorithm E (Exponential)/PD (Proportional Derivative)-Control. The first Event-Based control loop will be implemented to prevent sudden drop of the learning rate when the model is approaching the optimum. The second Event-Based control loop will decide, based on the learning speed, when to switch to the next data batch. Experimental evaluation is provided using two state-of-the-art machine learning image datasets (CIFAR-10 and CIFAR-100). Results show the Event-Based E/PD is better than the original algorithm (higher final accuracy, lower final loss value), and the Double-Event-Based E/PD can accelerate the training process, save up to 67% training time compared to state-of-the-art algorithms and even result in better performance.

This paper presents a novel urban transportation mode detection (TMD) method based on inertial and barometric data. To the best of the authors knowledge, this is the first time that a 3-axis accelerometer, a 3-axis gyroscope, and a barometer are combined for TMD purposes. One contribution was to build and share an optimized TMD dataset in terms of duration (44 hours), number of participants (34), sensor placements (on-foot, waist-attached and in the trouser’s pocket) and considered transportation modes (7). To infer transportation mode information, we tackle two classification scenarios: still, walk, bike, tramway, and bus, on the one hand, and still, walk, bike, elevator, stairs and public transport on the other hand. One major finding was that TMD classification accuracy has been widely overestimated by adopting the familiar randomized cross validation approach. According to the latter, the accuracy of our non-optimized classifiers ranged from 90% to 99%. However, few of them were accurate when the test samples were drawn from data of five unknown subjects. By selecting features in a personalized manner and by studying the importance of the placement of sensors, we realized that the latter significantly impact the performance of TMD models. Thus, we present here two robust TMD classifiers that show respectively an average accuracy of 75.63% for the first scenario and 79.41% for the second scenario using a foot-mounted sensor

Convolutional neural networks (CNNs) are commonly used for image classification tasks, raising the challenge of their application on data flows. During their training, adaptation is often performed by tuning the learning rate. Usual learning rate strategies are time-based i.e. monotonously decreasing. In this paper, we advocate switching to a performance-based adaptation, in order to improve the learning efficiency. We present E (Exponential)/PI (Proportional Integral)-Control, a conditional learning rate strategy that combines a feedback PI controller based on the CNN loss function, with an exponential control signal to smartly boost the learning and adapt the PI parameters. Stability proof is provided as well as an experimental evaluation using two state of the art image datasets (CIFAR-10 and Fashion-MNIST). Results show better performances than the related works (faster network accuracy growth reaching higher levels) and robustness of the E/PI-Control regarding its parametrization.

We consider the problem of active reduction of the structural vibrations induced by the sloshing of large masses of fuel inside a partly full tank. The proposed study focuses on an experimental device mimicking an aircraft wing made of an aluminium rectangular plate equipped with piezoelectric patches at the clamped end and with a cylindrical tip tank, more or less filled with liquid. After deriving a representative finite-dimensional model of the complete system, containing the first five structural modes of the plate and the first three liquid sloshing modes, a controller is computed. Since our main scope is to control the most energetic mode of the structure, a full state-feedback method coupled with an observer is used. Finally, the controller is also tested for different initial conditions/perturbations and the results are compared with the ones obtained with an H _∞ controller. Experimental results illustrate the relevance of the chosen strategy.