There's a lot of talk about the applicability of artificial intelligence (AI) and deep learning to taming the vast quantities of data that modern Operations teams and their tools deal with. Analyst reports frequently tout AI capabilities, no matter how minor, as a strength of a product, and the lack of them as a weakness. Yet no effective use of AI seems to have emerged and claimed wide adoption in Network Operations or Server Monitoring. Why not? (Disclaimer: LogicMonitor does not currently have deep learning or other AI capabilities). Part of the issue is that AI is a soft definition.

Suggestions, improvements and contributions are very much welcome!

Machines today can identify objects in images, but they are unable to fully decipher the most important aspect: what's actually happening in front of the camera. At TwentyBN, we have created the world's first AI technology that shows an awareness of its environment and of the actions occurring within it. Our system observes the world through live video and automatically interprets the unfolding visual scene. As with other technologies before it, deep learning has followed a series of step functions defined by sudden, often unexpected, outbreaks of capability. Each step function fundamentally pushed the envelope beyond what computers were previously able to achieve.

With machine learning receiving a significant amount of attention in finance, the UBS Global Quantitative research team wanted to recognize how deep learning, a related but more discerning aspect of artificial intelligence recognition, might benefit investors. To tackle the issue UBS called Matthew Dixon, Professor of Finance & Statistics at the Illinois Institute of Technology, to explain how very high dimensional input and hierarchical, data compression, multi-layer networks might benefit a stock portfolio. Dixon has experience working in the banking and high frequency trading in addition to holding a Ph.D. from the Imperial College London, all which points to the cutting edge of the cutting edge in computational finance. Get the entire 10-part series on Ray Dalio in PDF. Dixon had a primary goal in his conference call with UBS and their clients.

In this talk, we analyze the performance characteristics of Caffe* and TensorFlow* on Intel Xeon Phi processor x200. It is the latest processor using Intel Many Integrated Core Architecture (Intel MIC Architecture). It introduces several state-of-the-art features such as a compute core with two 512-bit vector processing units and an on-chip, high-bandwidth multichannel DRAM (MCDRAM) memory, delivering a theoretical peak performance of 6 TF single precision and 3 TF double precision floating point operations per second. We give an overview of the DNN framework architectures and describe the usage of Intel Math Kernel Library for Deep Neural Networks (Intel MKL-DNN) APIs in the implementation of different neural network layer computations. We present the details on the integration and performance optimizations of few of the compute intensive layers using Intel MKL-DNN APIs.

Since the creation of Generative Adversarial Networks (GANs), much work has been done to improve their training stability, their generated image quality, their range of application but nearly none of them explored their self-training potential. Self-training has been used before the advent of deep learning in order to allow training on limited labelled training data and has shown impressive results in semi-supervised learning. In this work, we combine these two ideas and make GANs self-trainable for semi-supervised learning tasks by exploiting their infinite data generation potential. Results show that using even the simplest form of self-training yields an improvement. We also show results for a more complex self-training scheme that performs at least as well as the basic self-training scheme but with significantly less data augmentation.

ABSTRACT In portable, 3-D, or ultra-fast ultrasound (US) imaging systems, there is an increasing demand to reconstruct high quality images from limited number of data. However, the existing solutions require either hardware changes or computationally expansive algorithms. To overcome these limitations, here we propose a novel deep learning approach that interpolates the missing RF data by utilizing the sparsity of the RF data in the Fourier domain. Extensive experimental results from sub-sampled RF data from a real US system confirmed that the proposed method can effectively reduce the data rate without sacrificing the image quality. Index Terms-- Deep learning, ultrasound imaging, lowrank Hankel matrix 1. INTRODUCTION Due to the the excellent temporal resolution with reasonable image quality and minimal invasiveness, ultrasound imaging has been adopted as a golden-standard for many disease diagnosis in heart, liver, etc. Accordingly, there have been many research efforts to extend the US imaging to new applications such as portable imaging in emergency care, 3-D imaging, ultra-fast imaging, etc.

The correlation length-scale next to the noise variance are the most used hyperparameters for the Gaussian processes. Typically, stationary covariance functions are used, which are only dependent on the distances between input points and thus invariant to the translations in the input space. The optimization of the hyperparameters is commonly done by maximizing the log marginal likelihood. This works quite well, if the distances are uniform distributed. In the case of a locally adapted or even sparse input space, the prediction of a test point can be worse dependent of its position. A possible solution to this, is the usage of a non-stationary covariance function, where the hyperparameters are calculated by a deep neural network. So that the correlation length scales and possibly the noise variance are dependent on the test point. Furthermore, different types of covariance functions are trained simultaneously, so that the Gaussian process prediction is an additive overlay of different covariance matrices. The right covariance functions combination and its hyperparameters are learned by the deep neural network. Additional, the Gaussian process will be able to be trained by batches or online and so it can handle arbitrarily large data sets. We call this framework Deep Gaussian Covariance Network (DGCP). There are also further extensions to this framework possible, for example sequentially dependent problems like time series or the local mixture of experts. The basic framework and some extension possibilities will be presented in this work. Moreover, a comparison to some recent state of the art surrogate model methods will be performed, also for a time dependent problem.

We propose a new approach to train the Generative Adversarial Nets (GANs) with a mixture of generators to overcome the mode collapsing problem. The main intuition is to employ multiple generators, instead of using a single one as in the original GAN. The idea is simple, yet proven to be extremely effective at covering diverse data modes, easily overcoming the mode collapse and delivering state-of-the-art results. A minimax formulation is able to establish among a classifier, a discriminator, and a set of generators in a similar spirit with GAN. Generators create samples that are intended to come from the same distribution as the training data, whilst the discriminator determines whether samples are true data or generated by generators, and the classifier specifies which generator a sample comes from. The distinguishing feature is that internal samples are created from multiple generators, and then one of them will be randomly selected as final output similar to the mechanism of a probabilistic mixture model. We term our method Mixture GAN (MGAN). We develop theoretical analysis to prove that, at the equilibrium, the Jensen-Shannon divergence (JSD) between the mixture of generators' distributions and the empirical data distribution is minimal, whilst the JSD among generators' distributions is maximal, hence effectively avoiding the mode collapse. By utilizing parameter sharing, our proposed model adds minimal computational cost to the standard GAN, and thus can also efficiently scale to large-scale datasets. We conduct extensive experiments on synthetic 2D data and natural image databases (CIFAR-10, STL-10 and ImageNet) to demonstrate the superior performance of our MGAN in achieving state-of-the-art Inception scores over latest baselines, generating diverse and appealing recognizable objects at different resolutions, and specializing in capturing different types of objects by generators.

ABSTRACT Long short-term memory (LSTM) is normally used in recurrent neural network (RNN) as basic recurrent unit. However, conventional LSTM assumes that the state at current time step depends on previous time step. In this study, we propose a new variation of LSTM, advanced LSTM (A-LSTM), for better temporal context modeling. We employ A-LSTM in weighted pooling RNN for emotion recognition. The A-LSTM outperforms the conventional LSTM by 5.5% relatively. The A-LSTM based weighted pooling RNN can also complement the state-of-the-art emotion classification framework. This shows the advantage of A-LSTM. Index Terms-- multi-task learning, attention model, long short-term memory, recurrent neural network, emotion recognition 1. INTRODUCTION Recurrent neural network is recently used as a dynamic model for sequential input.