Deep Neural Networks (DNNs) have revolutionized numerous applications, but the demand for ever more performance remains unabated. Scaling DNN computations to larger clusters is generally done by distributing tasks in batch mode using methods such as distributed synchronous SGD. Among the issues with this approach is that to make the distributed cluster work with high utilization, the workload distributed to each node must be large, which implies nontrivial growth in the SGD mini-batch size. In this paper, we propose a framework called FPDeep, which uses a hybrid of model and layer parallelism to configure distributed reconfigurable clusters to train DNNs. This approach has numerous benefits. First, the design does not suffer from batch size growth. Second, novel workload and weight partitioning leads to balanced loads of both among nodes. And third, the entire system is a fine-grained pipeline. This leads to high parallelism and utilization and also minimizes the time features need to be cached while waiting for back-propagation. As a result, storage demand is reduced to the point where only on-chip memory is used for the convolution layers. We evaluate FPDeep with the Alexnet, VGG-16, and VGG-19 benchmarks. Experimental results show that FPDeep has good scalability to a large number of FPGAs, with the limiting factor being the FPGA-to-FPGA bandwidth. With 6 transceivers per FPGA, FPDeep shows linearity up to 83 FPGAs. Energy efficiency is evaluated with respect to GOPs/J. FPDeep provides, on average, 6.36x higher energy efficiency than comparable GPU servers.

In this research work, we perform text line segmentation directly in compressed representation of an unconstrained handwritten document image. In this relation, we make use of text line terminal points which is the current state-of-the-art. The terminal points spotted along both margins (left and right) of a document image for every text line are considered as source and target respectively. The tunneling algorithm uses a single agent (or robot) to identify the coordinate positions in the compressed representation to perform text-line segmentation of the document. The agent starts at a source point and progressively tunnels a path routing in between two adjacent text lines and reaches the probable target. The agent's navigation path from source to the target bypassing obstacles, if any, results in segregating the two adjacent text lines. However, the target point would be known only when the agent reaches the destination; this is applicable for all source points and henceforth we could analyze the correspondence between source and target nodes. Artificial Intelligence in Expert systems, dynamic programming and greedy strategies are employed for every search space while tunneling. An exhaustive experimentation is carried out on various benchmark datasets including ICDAR13 and the performances are reported.

Abstract-- The present paper shows a solution to the problem of automatic distress detection, more precisely the detection of holes in paved roads. To do so, the proposed solution uses a weightless neural network known as Wisard to decide whether an image of a road has any kind of cracks. In addition, the proposed architecture also shows how the use of transfer learning was able to improve the overall accuracy of the decision system. As a verification step of the research, an experiment was carried out using images from the streets at the Federal University of Tocantins, Brazil. The architecture of the developed solution presents a result of 85.71% accuracy in the dataset, proving to be superior to approaches of the state-of-the-art. I.INTRODUCTION In Brazil, most of the traffic is driven on asphalt roads.

Abstract--In this work we investigate approaches to reconstruct generator models from measurements available at the generator terminal bus using machine learning (ML) techniques. The goal is to develop an emulator which is trained online and is capable of fast predictive computations. The training is illustrated on synthetic data generated based on available open-source dynamical generator model. Two ML techniques were developed and tested: (a) standard vector auto-regressive (VAR) model; and (b) novel customized long short-term memory (LSTM) deep learning model. Tradeoffs in reconstruction ability between computationally light but linear AR model and powerful but computationally demanding LSTM model are established and analyzed.

You might ask what the difference is between most artificial intelligence (AI) companies and SparkCognition. Here it is: while at other firms, humans build models; SparkCognition puts them together with algorithms. Rather than roughing out one model and then doing a bunch of testing, SparkCognition continually tests and fits models to data accumulating in real time, an architecture that allows it to deal with big data. Without foregone conclusions about what might be happening, SparkCognition algorithms keep probing for relationships and possible explanations without any a priori idea of what's going on. This fantastic flexibility, along with the speed of computer technology, allows SparkCognition to come to conclusions fast enough for real-time intervention.

Deep inside the Earth, miles down in many cases, rock-sealed pockets hold buried treasures. These hydrocarbon reservoirs are packed with organic compounds that make the world go'round. When the contents are extracted and refined, the resulting oil and gas help light cities, transport people and run industries. For some engineers at BP, Job One is locating the reservoirs. Job Two is accurately predicting what percentage of hydrocarbons are retrievable, also known as "recovery factor."

Smart cities can add value not just through money saved by efficient operations but also through the added benefits for citizens and local businesses by adding great design. The key to building truly "smart" and efficient cities of the future is improving not just infrastructure -- but the general makeup of its entire foundation. This includes concepts like energy and resource management, traffic and productivity, public services, active culture, and education. These few "extras" are definitely needed on a broad scale. It's all of these things merged to create a well-oiled, highly efficient machine.

Low-rank matrix factorization (LRMF) has received much popularity owing to its successful applications in both computer vision and data mining. By assuming the noise term to come from a Gaussian, Laplace or a mixture of Gaussian distributions, significant efforts have been made on optimizing the (weighted) $L_1$ or $L_2$-norm loss between an observed matrix and its bilinear factorization. However, the type of noise distribution is generally unknown in real applications and inappropriate assumptions will inevitably deteriorate the behavior of LRMF. On the other hand, real data are often corrupted by skew rather than symmetric noise. To tackle this problem, this paper presents a novel LRMF model called AQ-LRMF by modeling noise with a mixture of asymmetric Laplace distributions. An efficient algorithm based on the expectation-maximization (EM) algorithm is also offered to estimate the parameters involved in AQ-LRMF. The AQ-LRMF model possesses the advantage that it can approximate noise well no matter whether the real noise is symmetric or skew. The core idea of AQ-LRMF lies in solving a weighted $L_1$ problem with weights being learned from data. The experiments conducted with synthetic and real datasets show that AQ-LRMF outperforms several state-of-the-art techniques. Furthermore, AQ-LRMF also has the superiority over the other algorithms that it can capture local structural information contained in real images.

Multi-output learning aims to simultaneously predict multiple outputs given an input. It is an important learning problem due to the pressing need for sophisticated decision making in real-world applications. Inspired by big data, the 4Vs characteristics of multi-output imposes a set of challenges to multi-output learning, in terms of the volume, velocity, variety and veracity of the outputs. Increasing number of works in the literature have been devoted to the study of multi-output learning and the development of novel approaches for addressing the challenges encountered. However, it lacks a comprehensive overview on different types of challenges of multi-output learning brought by the characteristics of the multiple outputs and the techniques proposed to overcome the challenges. This paper thus attempts to fill in this gap to provide a comprehensive review on this area. We first introduce different stages of the life cycle of the output labels. Then we present the paradigm on multi-output learning, including its myriads of output structures, definitions of its different sub-problems, model evaluation metrics and popular data repositories used in the study. Subsequently, we review a number of state-of-the-art multi-output learning methods, which are categorized based on the challenges.

Mini-robots that can fix pipe leaks and work in dangerous environments could soon make road-works a thing of the past. The government is investing a total of £26.6 million to help fund a variety of projects, including the 1cm-long machines. The cash is being split among four UK universities - Leeds, Sheffield, Birmingham and Bristol - to create the innovative pipe-bots which will work in underground pipes and dangerous sites like nuclear decommissioning facilities. It is hoped they will spell the end for many disruptive and expensive roadworks, as robots carry out repairs without the need to dig up the roads. Mini-robots that can fix pipe leaks and work in dangerous environments could soon make road-works a thing of the past.