In this blog series, I've focused on how NetApp can help you streamline your artificial intelligence projects. With technologies and services for managing data everywhere, NetApp is well positioned to solve your AI data challenges. Built on our partnership with NVIDIA and powered by NVIDIA DGX supercomputers and NetApp all-flash storage, ONTAP AI lets you simplify, accelerate, and scale your AI data pipeline to gain deeper understanding in less time. Combining Data Fabric enabled NetApp storage with GPU-accelerated NVIDIA computing systems results in capabilities that aren't available from other turnkey AI solutions, on-premises or in the cloud. Here are five of the key advantages of ONTAP AI.

Three institutions working together have applied DeepMind's neural network learning system to the task of discovering and diagnosing eye diseases. Moorfields Eye Hospital has been working with Google's DeepMind Health subsidiary and University College London in the effort, and have documented their progress in a paper published in Nature Medicine. As the researchers note, eye doctors currently use a machine that carries out optical coherence tomography (OCT) on patients to find out if they have an eye disease. While the technique is quite useful and accurate, it requires highly trained doctors to spend time looking at results. The researchers suggest this creates a backlog that sometimes prevents patients from getting the care they need in time to save their vision.

In giving the diagnosis, the doctor combines the information obtained through such processes with his or her own knowledge and experience. No matter how reputed a physician may be, the chances of them making a wrong diagnosis can never be zero. With the recent progress in artificial intelligence, there has been much speculation that artificial intelligence could very well surpass a human doctor's ability to make diagnoses and write prescriptions. In August 2016, the Institute of Medical Science at the University of Tokyo released the outcome of a case study to show how powerful AI can be. IBM's Watson AI program was fed with information contained in nearly 20 million medical articles related to cancer research and more than 17 million pieces of information related to pharmaceuticals.

Despite the successful application of machine learning (ML) in a wide range of domains, adaptability---the very property that makes machine learning desirable---can be exploited by adversaries to contaminate training and evade classification. In this paper, we investigate the feasibility of applying a specific class of machine learning algorithms, namely, reinforcement learning (RL) algorithms, for autonomous cyber defence in software-defined networking (SDN). In particular, we focus on how an RL agent reacts towards different forms of causative attacks that poison its training process, including indiscriminate and targeted, white-box and black-box attacks. In addition, we also study the impact of the attack timing, and explore potential countermeasures such as adversarial training.

Deep neuroevolution, that is evolutionary policy search methods based on deep neural networks, have recently emerged as a competitor to deep reinforcement learning algorithms due to their better parallelization capabilities. However, these methods still suffer from a far worse sample efficiency. In this paper we investigate whether a mechanism known as "importance mixing" can significantly improve their sample efficiency. We provide a didactic presentation of importance mixing and we explain how it can be extended to reuse more samples. Then, from an empirical comparison based on a simple benchmark, we show that, though it actually provides better sample efficiency, it is still far from the sample efficiency of deep reinforcement learning, though it is more stable.

This paper describes the Pressure Ulcers Online Website, which is a first step solution towards a new and innovative platform for helping people to detect, understand and manage pressure ulcers. It outlines the reasons why the project has been developed and provides a central point of contact for pressure ulcer analysis and ongoing research. Using state-of-the-art technologies in convolutional neural networks and transfer learning along with end-to-end web technologies, this platform allows pressure ulcers to be analysed and findings to be reported. As the system evolves through collaborative partnerships, future versions will provide decision support functions to describe the complex characteristics of pressure ulcers along with information on wound care across multiple user boundaries. This project is therefore intended to raise awareness and support for people suffering with or providing care for pressure ulcers.

Recent algorithmic improvements and hardware breakthroughs resulted in a number of success stories in the field of AI impacting our daily lives. However, despite its ubiquity AI is only just starting to make advances in what may arguably have the largest impact thus far, the nascent field of autonomous driving. In this work we discuss this important topic and address one of crucial aspects of the emerging area, the problem of predicting future state of autonomous vehicle's surrounding necessary for safe and efficient operations. We introduce a deep learning-based approach that takes into account current state of traffic actors and produces rasterized representations of each actor's vicinity. The raster images are then used by deep convolutional models to infer future movement of actors while accounting for inherent uncertainty of the prediction task. Extensive experiments on real-world data strongly suggest benefits of the proposed approach. Moreover, following successful tests the system was deployed to a fleet of autonomous vehicles.

This problem is known as phase retrieval and is encountered frequently in applications including X-ray crystallography [1, 2], astronomy [3], optics [4], tomography, microscopy, array imaging [5], acoustics [6], quantum mechanics [7] and ptychography [8], where it is extremely difficult or infeasible to measure phase information of signal while recording magnitude measurements is much easier. In its full generality, the inverse problem 1 is severely ill-posed due to its nonlinear and non-convex nature. Traditional approaches to overcome the ill posedness of phase retrieval generally falls into two categories. First approach is to introduce redundancy into measurement system, where we take more measurements than dimension of true signal x, i.e., m n usually in the form of oversampled Fourier transform [9], short-time Fourier transform [10], random Gaussian measurements [11], coded diffraction patterns using random masks or structured illuminations [12, 13], wavelet transform [14], and Gabor frames [15]. Second approach is to exploit some known knowledge about true signal x (prior information) such as sparsity [16, 17, 18] or non-negativity [19, 20].

Nowadays, most of the objective speech quality assessment tools (e.g., perceptual evaluation of speech quality (PESQ)) are based on the comparison of the degraded/processed speech with its clean counterpart. The need of a "golden" reference considerably restricts the practicality of such assessment tools in real-world scenarios since the clean reference usually cannot be accessed. On the other hand, human beings can readily evaluate the speech quality without any reference (e.g., mean opinion score (MOS) tests), implying the existence of an objective and non-intrusive (no clean reference needed) quality assessment mechanism. In this study, we propose a novel end-to-end, non-intrusive speech quality evaluation model, termed Quality-Net, based on bidirectional long short-term memory. The evaluation of utterance-level quality in Quality-Net is based on the frame-level assessment. Frame constraints and sensible initializations of forget gate biases are applied to learn meaningful frame-level quality assessment from the utterance-level quality label. Experimental results show that Quality-Net can yield high correlation to PESQ (0.9 for the noisy speech and 0.84 for the speech processed by speech enhancement). We believe that Quality-Net has potential to be used in a wide variety of applications of speech signal processing.

The performance of Feedforward neural network (FNN) fully depends upon the selection of architecture and training algorithm. FNN architecture can be tweaked using several parameters, such as the number of hidden layers, number of hidden neurons at each hidden layer and number of connections between layers. There may be exponential combinations for these architectural attributes which may be unmanageable manually, so it requires an algorithm which can automatically design an optimal architecture with high generalization ability. Numerous optimization algorithms have been utilized for FNN architecture determination. This paper proposes a new methodology which can work on the estimation of hidden layers and their respective neurons for FNN. This work combines the advantages of Tabu search (TS) and Gradient descent with momentum backpropagation (GDM) training algorithm to demonstrate how Tabu search can automatically select the best architecture from the populated architectures based on minimum testing error criteria. The proposed approach has been tested on four classification benchmark dataset of different size Keywords: Tabu search (TS), Feedforward neural network (FNN), hidden layer, hidden neurons, optimization, architecture.