Although kernel methods such as support vector machines are one of the state-of-the-art methods when it comes to fully automated learning, see e.g. the recent independent comparison [7], the recent years have shown that on complex datasets such as image, speech and video data, they clearly fall short compared to deep neural networks. One possible explanation for this superior behavior is certainly their deep architecture that makes it possible to represent highly complex functions with relatively few parameters. In particular, it is possible to amplify or suppress certain dimensions or features of the input data, or to combine features to new, more abstract features. Compared to this, standard kernels such as the popular Gaussian kernels simply treat every feature equally. In addition, most users of kernel machines probably stick to the very few standard kernels, often simply because there is in most cases no principled way for finding problem specific kernels.

Bedside monitors in Intensive Care Units (ICUs) frequently sound incorrectly, slowing response times and desensitising nurses to alarms (Chambrin, 2001), causing true alarms to be missed (Hug et al., 2011). We compare sliding window predictors with recurrent predictors to classify patient state-of-health from ICU multivariate time series; we report slightly improved performance for the RNN for three out of four targets.

The field of pathology diagnosis has steadily advanced with the development of microscopy, accompanied by the automation of the reduction of inter-observer reliability and intra-observer reproducibility. Within the field of mammography, computer vision, and artificial intelligence (AI) techniques have been successfully applied to detect and characterize abnormalities of medical images [Winsberg et al., 1967; Ravdin et al., 2001]. This has resulted in a situation such that automated detection techniques can now implement an entire medical procedure with a high degree of accuracy. In addition, advances in computer hardware and software have increased the performance and reliability of parallel computing. The advances in this technology have, in turn, provided hardware and software advancements that are sufficiently robust to support the large computational requirements of complex Artificial Intelligence (AI) algorithms and their application to machine learning.

ABSTRACT Kernel-based K-means clustering has gained popularity due to its simplicity and the power of its implicit nonlinear representation of the data. A dominant concern is the memory requirement since memory scales as the square of the number of data points. We provide a new analysis of a class of approximate kernel methods that have more modest memory requirements, and propose a specific one-pass randomized kernel approximation followed by standard K-means on the transformed data. The analysis and experiments suggest the method is accurate, while requiring drastically less memory than standard kernel K-means and significantly less memory than Nyström based approximations. Index Terms-- Kernel methods, Unsupervised learning, Lowrank approximation, Randomized algorithm 1. INTRODUCTION Kernel-based approaches are popular methods for supervised and unsupervised learning [1].

Accurate and robust cell nuclei classification is the cornerstone for a wider range of tasks in digital and Computational Pathology. However, most machine learning systems require extensive labeling from expert pathologists for each individual problem at hand, with no or limited abilities for knowledge transfer between datasets and organ sites. In this paper we implement and evaluate a variety of deep neural network models and model ensembles for nuclei classification in renal cell cancer (RCC) and prostate cancer (PCa). We propose a convolutional neural network system based on residual learning which significantly improves over the state-of-the-art in cell nuclei classification. Finally, we show that the combination of tissue types during training increases not only classification accuracy but also overall survival analysis.

We propose a semidefinite programming (SDP) algorithm for community detection in the stochastic block model, a popular model for networks with latent community structure. We prove that our algorithm achieves exact recovery of the latent communities, up to the information-theoretic limits determined by Abbe and Sandon (2015). Our result extends prior SDP approaches by allowing for many communities of different sizes. By virtue of a semidefinite approach, our algorithms succeed against a semirandom variant of the stochastic block model, guaranteeing a form of robustness and generalization. We further explore how semirandom models can lend insight into both the strengths and limitations of SDPs in this setting.

The Witcher 3 doesn't really belong in a 2016 video game award ceremony. I'm excited to watch the 2016 Game Awards this evening. We're in for a few big game reveals and some gameplay footage for highly anticipated titles launching next year. That's all well and good, but I have a quibble with two of the nominees for Best RPG in this year's awards. Notice that here we have both The Witcher 3 and World of Warcaft occupying two of the five slots.

Machine learning is a powerful tool used for a variety of tasks in modern life, from fraud detection and sorting spam in Google, to making movie recommendations on Netflix. Now a team of researchers from the University of Toronto Scarborough have developed a novel approach in using it to determine whether planetary systems are stable or not. "Machine learning offers a powerful way to tackle a problem in astrophysics, and that's predicting whether planetary systems are stable," says Dan Tamayo, lead author of the research and a postdoctoral fellow in the Centre for Planetary Science at U of T Scarborough. Machine learning is a form of artificial intelligence that gives computers the ability to learn without having to be constantly programmed for a specific task. The benefit is that it can teach computers to learn and change when exposed to new data, not to mention it's also very efficient.

"Our research has demonstrated for the first time that machine-learning algorithms can assist in the discrimination of physiological versus pathological hypertrophic remodeling, thus enabling easier and more accurate diagnoses of HCM," said senior study author Partho P. Sengupta, MD, Director of Cardiac Ultrasound Research and Professor of Medicine in Cardiology at the Icahn School of Medicine at Mount Sinai.

Amazon is partnering up with Techstars to launch an Alexa Accelerator for companies developing conversational artificial intelligence. Starting in January 2017, early-stage companies working on speech technology or voice applications in any industry--including, but not limited to automobiles, health, communications, wearables, and connected homes--can apply for the 13-week program, and a chance to win $120,000. Ten to 12 startups will have a spot in the program, and will live at the University of Washington beginning in July 2017. Participants will each receive a total of $20,000, and will have a chance to work with Amazon and Techstars mentors to develop strategies that will help further their business. A demo day will be held in October during which companies will share their product with investors, and a select few will win an additional $100,000 at the conclusion of the program.