In the aftermath of disasters, building damage maps are ob-tained using change detection to plan rescue operations. Cur-rent convolutional neural network approaches do not consider the similarities between neighboring buildings for predicting the damage. We present a novel graph-based building dam-age detection solution to capture these relationships. Our pro-posed model architecture learns from both local and neigh-borhood features to predict building damage. Specifically, we adopt the sample and aggregate graph convolution strategy to learn aggregation functions that generalize to unseen graphs which is essential for alleviating the time needed to obtain predictions for new disasters. Our experiments on the xBD dataset and comparisons with a classical convolutional neu-ral network reveal that while our approach is handicapped by class imbalance, it presents a promising and distinct advan-tage when it comes to cross-disaster generalization.

MOTIVATION: Identifying histone tail modifications using ChIP-seq is commonly used in time-series experiments in development and disease. These assays, however, cover specific time-points leaving intermediate or early stages with missing information. Although several machine learning methods were developed to predict histone marks, none exploited the dependence that exists in time-series experiments between data generated at specific time-points to extrapolate these findings to time-points where data cannot be generated for lack or scarcity of materials (i.e. early developmental stages). RESULTS: Here, we train a deep learning model named TempoMAGE, to predict the presence or absence of H3K27ac in open chromatin regions by integrating information from sequence, gene expression, chromatin accessibility and the estimated change in H3K27ac state from a reference time-point. We show that adding reference time-point information systematically improves the overall model's performance. In addition, sequence signatures extracted from our method were exclusive to the training dataset indicating that our model learned data-specific features. As an application, TempoMAGE was able to predict the activity of enhancers from pre-validated in-vivo dataset highlighting its ability to be used for functional annotation of putative enhancers. AVAILABILITY AND IMPLEMENTATION: TempoMAGE is freely available through GitHub at https://github.com/pkhoueiry/TempoMAGE. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Research has shown that deep neural networks are able to help and assist human workers throughout the industrial sector via different computer vision applications. However, such data-driven learning approaches require a very large number of labeled training images in order to generalize well and achieve high accuracies that meet industry standards. Gathering and labeling large amounts of images is both expensive and time consuming, specifically for industrial use-cases. In this work, we introduce BAR (Bounding-box Automated Refinement), a reinforcement learning agent that learns to correct inaccurate bounding-boxes that are weakly generated by certain detection methods, or wrongly annotated by a human, using either an offline training method with Deep Reinforcement Learning (BAR-DRL), or an online one using Contextual Bandits (BAR-CB). Our agent limits the human intervention to correcting or verifying a subset of bounding-boxes instead of re-drawing new ones. Results on a car industry-related dataset and on the PASCAL VOC dataset show a consistent increase of up to 0.28 in the Intersection-over-Union of bounding-boxes with their desired ground-truths, while saving 30%-82% of human intervention time in either correcting or re-drawing inaccurate proposals.

This paper describes the methodology used to design a NOxpredictive emissions monitoring system (PEMS) based on an artificial neural network (ANN). To find the optimal ANN architecture, a QGA-based search was performed over the set of possible model architectures, the activation function in each layer and the learning rate were architecture parameters taken into consideration. In addition, the QGA performed feature selection to remove non-significant input parameters that did not contribute significantly to the output. The objective function included penalties for network complexity and generalization error, among which a newly introduced penalty that makes use of the effective number of parameters provided by Bayesian Regularization. The developed framework was tested on data collected from a power plant consisting of twelve diesel engine-powered generators. It was found that the newly introduced penalty was enough to yield well-performing ANN's who demonstrated superior performance when compared to some traditional ML models, with up to a 75% reduction in the mean relative error when compared with a normal radial-basis network. In addition, the QGA was better at avoiding local optima, and on average converged in half the time required for the classical GA to converge.

Have you ever wondered how a song might sound if performed by a different artist? In this work, we propose SCM-GAN, an end-to-end non-parallel song conversion system powered by generative adversarial and transfer learning, which allows users to listen to a selected target singer singing any song. SCM-GAN first separates songs into vocals and instrumental music using a U-Net network, then converts the vocal segments to the target singer using advanced CycleGAN-VC, before merging the converted vocals with their corresponding background music. SCM-GAN is first initialized with feature representations learned from a state-of-the-art voice-to-voice conversion and then trained on a dataset of non-parallel songs. After that, SCM-GAN is evaluated against a set of metrics including global variance GV and modulation spectra MS on the 24 Mel-cepstral coefficients (MCEPs). Transfer learning improves the GV by 35% and the MS by 13% on average. A subjective comparison is conducted to test the output's similarity to the target singer and its naturalness. Results show that the SCM-GAN's similarity between its output and the target reaches 69%, and its naturalness reaches 54%.

In this work, we propose a novel Generative Adversarial Stacked Autoencoder that learns to map facial expressions with up to ±60 degrees to an illumination invariant facial representation of 0 degrees. We accomplish this by using a novel convolutional layer that exploits both local and global spatial information, and a convolutional layer with a reduced number of parameters that exploits facial symmetry. Furthermore, we introduce a generative adversarial gradual greedy layer-wise learning algorithm designed to train Adversarial Autoencoders in an efficient and incremental manner. We demonstrate the efficiency of our method and report state-of-the-art performance on several facial emotion recognition corpora, including one collected in the wild.

Bright spots, strong indicators of the existence of hydrocarbon accumulations, have been primarily used by geophysicists in oil and gas exploration. Recently, machine-learning algorithms, adopted to automate bright spot detection, have mainly relied on feature extraction and shallow classification workflows to achieve an 85.4% F1 score at best, on 2-D seismic data. Deep neural networks have proved their effectiveness in image classification applications, outperforming humans in some instances, but have not been applied to bright spot detection yet. However, their data-hungry nature poses a challenge in domains suffering from expensive data acquisition, such as seismic data analysis problems; they generally require millions of training samples before achieving good performance. In this article, we implement SeisNet, a convolutional neural network with a “butterfly” architecture that overcame the limited data challenge by implementing data augmentation and inductive transfer-learning techniques. Moreover, we adopt a novel formulation that allows us to detect bright spots and estimate their volume. Our approach was tested against different pretraining and transfer-learning methods and was shown to outperform other approaches in the literature by achieving a 95.6% F1 score on bright spot detection. Our model accurately predicted the volume of the bright spot with an average absolute error that is not more than 0.04% of the total volume of the seismic image. This article is an important step in establishing pretrained networks for other seismic applications such as earthquake prediction; our domain-specific pretrained network, proven to outperform state-of-the-art pretrained networks on bright spot detection, may be used to jump-start the training of deep models on other seismic problems.

Medical students undergo exams, called "Objective Structured Clinical Examinations" (OSCEs), to assess their medical competence in clinical tasks. In these OSCEs, a medical student interacts with a standardized patient, asking questions to complete a clinical assessment of the patient's medical case. In real OSCEs, standardized patients or "Actors" are recruited and trained to answer questions about symptoms mentioned in a script designed by the medical examiner. Developing a virtual conversational patient for OSCEs would lead to significant logistical savings. In this work, we develop a deep learning framework to improve the virtual patient's conversational skills. First, deep neural networks learned domain specific word embeddings. Then, long short-term memory networks derived sentence embeddings before a convolutional neural network model selected an answer to a given question from a script. Empirical results on a homegrown corpus showed that this framework outperformed other approaches, and reached an accuracy of 81%.

With the “last mile” of the delivery process being the most expensive phase, autonomous package delivery systems are gaining traction as they aim for faster and cheaper delivery of goods to city, urban and rural destinations. This interest is further fueled by the emergence of e-commerce, where many applications can benefit from autonomous package delivery solutions. However, the environment stochasticity, variability and task complexity for autonomous operation make it difficult to deploy such systems in real-world applications without the incorporation of advanced machine learning and optimization algorithms. Moving away from designing a “one size fits all” agent to solve the outdoor package delivery problem and considering ad-hoc teams of agents trained within a data-driven framework could provide the answer. In this work, we propose a delivery scheduling algorithm for heterogeneous multi-agent systems using the pickup and delivery problem (PDP) formulation. Specifically, a 3-index mixed integer program-based PDP that allows coalition formation (PDP-CF) among agents is derived to allow multi-agent PDP schedules. We propose a quantum genetic algorithm to solve for the schedule since it is can better handle the large computational complexity of PDP-CF. Multiple PDP scenario simulations show the merits of the proposed approach.

We introduce in this paper two main approaches, Triangulateration (TL) and Angle-Difference-of-Arrival (ADoA) for indoor localization and mapping using single-anchor and millimeter wave (MMW) propagation characteristics. Then, we perform context inference through obstacle localization. To do so, we first include and estimate the positions of virtual anchor nodes (VANs), known as mirrors of the real anchor with respect to obstacle. Then, it is followed by estimating the obstacle position and its dimensions. We assess the performance of each technique via cumulative distribution function (CDF) for the location estimation root mean square error (RMSE). Simulations confirm that localization of the receiver relying on a single anchor and the localization of obstacles in MMW achieves a few centimeters accuracy.

In a previous work, we presented a system which combines active solid state drives and reconfigurable FPGAs (which we called reconfigurable active SSD nodes, or simply RASSD nodes) into a storage-compute node that can be used by a cloud data-center to achieve accelerated computations while running data intensive applications. To hide the complexity of accessing RASSD nodes from applications, we proposed in another work a middleware framework which handles all low-level interactions with the hardware. The Middleware Server (MWS), which manages a group of RASSD nodes, has the role of bridging the connection between a client and the nodes. In this paper, we present extensions to the MWS to enable it to operate within a collaborative cloud environment, and we develop a model to evaluate the performance of the collaborative MWS. This model represents a study of the utilization of three hardware resources of the MWS: CPU, memory, and network interface. For each, we derive the parameters that affect its operations, and propose formulas for its utilization. The results describe the capacity of a MWS, and hence can be used to decide on the number of MWSs in a collaborative cloud datacenter.

A principled approach to machine learning (ML) problems because of its mathematical foundations in statistical learning theory, support vector machines (SVM), a non-parametric method, require all the data to be available during the training phase. However, once the model parameters are identified, SVM relies only, for future prediction, on a subset of these training instances, called support vectors (SV). The SVM model is mathematically written as a weighted sum of these SV whose number, rather than the dimensionality of the input space, defines SVM's complexity. Since the final number of these SV can be up to half the size of the training dataset, SVM becomes challenged to run on energy aware computing platforms. We propose in this work Knee-Cut SVM (KCSVM) and Knee-Cut Ordinal Optimization inspired SVM (KCOOSVM) that use a soft trick of ordered kernel values and uniform subsampling to reduce SVM's prediction computational complexity while maintaining an acceptable impact on its generalization capability. When tested on several databases from UCL KCSVM and KCOOSVM produced promising results, comparable to similar published algorithms.

In this article, we propose new extensions to Hadoop to enable clusters of reconfigurable active solid-state drives (RASSDs) to process streaming data from SSDs using FPGAs. We also develop an analytical model to estimate the performance of RASSD clusters running under Hadoop. Using the Hadoop RASSD platform and network simulators, we validate our design and demonstrate its impact on performance for different workloads taken from Stanford's Phoenix MapReduce project. Our results show that for a hardware acceleration factor of 20×, compute-intensive workloads processing 153MB of data can run up to 11× faster than a standard Hadoop cluster.

A novel and rapidly growing area of research concerns data-intensive applications and the technical challenges that accompany it. One of those challenges is developing approaches and mechanisms that render high performance in processing and storing data. We joined this research effort by proposing a reconfigurable active solid state drives (RASSD) system that deals with such applications, through employing basic hardware, namely FPGA's connected to SSD's, to service the above applications as processing nodes, and take advantage of the close proximity between storage and processing. In this paper, we propose an intelligent middleware system for interfacing workstation-based and mobile applications to the distributed RASSD system. In order to provide high performance in terms of time and functionality, the middleware manages the data processing on the RASSD nodes through special pieces of code that we call drivelets, along with FPGA configuration files (bitstreams). Another important responsibility of the proposed middleware architecture lies in the unguided management of applications' flows, where it uses an intelligent script-parsing mechanism to turn one general request from the client into a sequence of operations needed to generate the required results. The middleware design allows for the integration of mobile applications into the overall architecture of the RASSD system, and allowing them to run data intensive applications that otherwise it is unfeasible for them to execute. We validate our design by comparing it to an existing middleware architecture, and present two use-cases with their results and discussion.

Class imbalance (CI) is common in most non synthetic datasets, which presents a major challenge for many classification algorithms geared towards optimized overall accuracy whenever the minority class risk loss is often higher than the majority class one. Support vector machine (SVM), a machine learning (ML) technique deeply rooted in statistics, maximizes linear margins between classes and generalizes well on yet to be seen data as long as the dataset is not severely imbalanced. Motivated to improve classification of imbalanced datasets using SVM standard formulation, we propose in this study a novel minority SVM (MinSVM) that achieves, with the addition of one constraint to the SVM objective function, separating boundaries that are closer to the majority class. Consequently, the minority data points are favored, and the probability of being misclassified becomes smaller. Experimental results support MinSVM claims and motivate to follow on research.

Cortical algorithms (CA) inspired by and modeled after the human cortex, have shown superior accuracy in few machine learning applications. However, CA have not been extensively implemented for speech recognition applications, in particular the Arabic language. Motivated to apply CA to Arabic speech recognition, we present in this paper an improved CA that is efficiently trained using an entropy-based cost function, and an entropy based weight update rule. We modify the strengthening and inhibiting rules originally employed in CA during feedback training with weighted entropy concepts. Preliminary results show the merit of the proposed modifications in the recognition of isolated Arabic speech and motivate follow on research.

Despite support vector machines' (SVM) robustness and optimality, SVM do not scale well computationally. Suffering from slow training convergence on large datasets, SVM online testing time can be suboptimal because SVM write the classifier hyper-plane model as a sum of support vectors that could total as much as half the datasets. Motivated to speed up SVM real time testing by reducing the number of SV, we introduce in this paper a novel local mixture based SVM (LMSVM) approach that exploits the increased separability provided by the kernel trick, while introducing a onetime computational expense. LMSVM applies kernel k-means clustering to the data in kernel space before pruning unwanted clusters based on a mixture measure for label heterogeneity. LMSVM's computational complexity and classification accuracy on four databases from UCI show promising results and motivate follow on research.

Capacitor Placement (CP) is a complex non linear mixed integer optimization problem that includes both integer variables that correspond to the locations at which capacitors should be placed and discrete variables that represent the number of capacitor banks that should be installed. Motivated to efficiently reduce power losses in distribution networks by CP and network reconfiguration (NR) using some of the intelligent soft computing approaches, we propose to use in this study ordinal optimization (OO) which relies on order comparison and goal softening to make the computation more efficient. We tested OO on 33 and 69 bus systems and compared experimental results with other recently published methods. The reduction in losses obtained using OO for CP and NR motivates further studies.

Load forecasting is a critical necessity in the electricity industry since any unanticipated demand could cause possible grid instability and blackouts. Ideally, the capacity should be kept slightly above the current demand to avoid any undesired outages and suboptimal last minute power purchase. Motivated to develop an intelligent and efficient forecasting approach, we propose investigating in this paper the impact of using a loss function in Support Vector Regression (SVR) that is modified with a strict mandate to minimize under estimating power needs. Experimental results for the municipality of Beirut's power substations show that the number of under-predictions was drastically reduced from an average of 50% to 1.91% with a very minimal impact of 0.3% on average on the error rate which motivates follow on research.

In this paper, we propose to combine active solid state drives and reconfigurable FPGAs into a storage-compute node to use as a building block in a distributed, high performance computation platform for data intensive applications. We propose a complete framework for middleware functionality through an API abstraction layer that hides the complexity of accessing and processing data stored on these distributed nodes, thus allowing programmers to focus on the application, and not the underlying specialized architecture. The application in turn is re-architected to maximize its performance by delegating selected computations down to the storage-compute node. We present preliminary results measured on a real hardware prototype of a single-node. These results show that our proposed architecture provides more than a 2× improvement in performance over non-reconfigurable active-disk architectures that use electromechanical disks for storage and a 6× improvement in performance over a platform that performs the computation on the middle server.

Given the exploding number of mobile subscribers and the quality of service requirements of emerging mobile services specially the multimedia ones, efficient use of the limited network resources has become so important that it is driving development of intelligent resource allocation (RA) algorithms. Whereas link specific RA problems have been widely investigated in the literature targeting either the uplink or downlink of a communication system, we present in this paper, SIRA, a novel joint uplink/downlink RA scenario in an OFDMA system for users running symmetrical services on their mobile devices. The RA process is modeled as a social gaming paradigm where users are considered as players interacting with the base station which takes the role of a principal controlling the RA game. Intelligently evolving from a soft RA and inspired by the model of team development proposed by Bruce Tuckman, SIRA uses penalty based concepts which play a feedback mechanism role from a system theory perspective to finalize the decision on allocated resources. Simulation results show the merit of the proposed technique and motivate follow on research.

Field Programmable Gate Arrays (FPGAs) inherent reconfigurable nature and their low power consumption have made them so complementary to microprocessors that many are advocating their inclusion in all supercomputing clusters. Today FPGAs are included in few mainstream computer systems for accelerating application specific performance. Among the numerous areas in reconfigurable computing FPGA have been encroaching into, we focus our literature review mainly on the area of high performance computing. Moving from FPGA general features to the evolution of FPGA supercomputing architecture, its roadmap, we reference selected applications lately developed for accelerating large simulation tasks using FPGA based supercomputers before presenting concluding remarks on challenges yet to be overcome.

This paper deals with a 3D methodology for brain tumor image-guided surgery. The methodology is based on development of a visualization process that mimics the human surgeon behavior and decision-making. In particular, it originally constructs a 3D representation of a tumor by using the segmented version of the 2D MRI images. Then it develops an optimal path for the tumor extraction based on minimizing the surgical effort and penetration area. A cost function, incorporated in this process, minimizes the damage surrounding healthy tissues taking into consideration the constraints of a new snake-like surgical tool proposed here. The tumor extraction method presented in this paper is compared with the ordinary method used on brain surgery, which is based on a straight-line based surgical tool. Illustrative examples based on real simulations present the advantages of the 3D methodology proposed here.