Big Data and the IoT are quickly transforming the world of clinical research, including how trial sponsors find and retain patients. While the digital revolution has permeated the medical world more slowly than other industries, it's finally begun to make a real impact. PwC research found that while the healthcare industry has a relatively low "Digital IQ" score of 65%, it boasts more CEOs that actively champion digital than any other industry. For clinical trials, that advocacy is translating into the implementation of big data and the Internet of Things (IoT), which are transforming not only how research is conducted, but the strategies used to identify, attract, and retain qualified patients. As methods of data collection and analysis become more sophisticated, clinical trial sponsors stand to make unprecedented progress in patient recruitment and retention.

According to Elo's research, the biggest threat for the well-being of Finnish entrepreneurs seem to be the challenge of time management and the lack of supporting networks. This is one of the challenges hackers in'Elo Data Science Hack' hackathon are set to solve. "The aim of our hackathon is to help entrepreneurs flourish and maintain balance in their working life. As Finland's most popular employment pension insurer, we feel passionate about Finnish entrepreneurs' well-being. We have over 85,000 customers and the only way to reach each and every one of them is with digital solutions. This also calls for a new kind of intelligent services," tells Satu Huber, CEO of Elo Mutual Pension Insurance Company.

The analysis has identified nearly 200 real-world enterprise use cases for AI that are classified into 25 industry sectors. The firm forecasts that revenue for enterprise AI applications will increase from 358 million in 2016 to 31.2 billion by 2025, representing a compound annual growth rate (CAGR) of 64.3%. Artificial intelligence (AI) technologies are quickly gaining mindshare among corporate executives around the world, driving a proliferation of use cases that touch virtually every industry. AI technologies, which include deep learning, machine learning, natural language processing (NLP), and computer vision, among others, are designed to endow computers with human-like faculties such as hearing, seeing, reasoning, and learning. But AI enables computers to do some things better than humans, especially when it comes to processing very large amounts of data quickly, efficiently, and accurately.

Over the summer, I spent some time in Silicon Valley, the home of technology and innovation. I took part in a program at Singularity University, a group whose mission is to help all of us understand how to utilize cutting-edge technologies to positively impact the world around us. I spend every day talking and thinking about technology and ways to make it do more for UBS Wealth Management's clients and staff. For two decades I've worked with different technology and I've seen some amazing changes – I remember not only when we didn't have emails on our phones in our pockets, but when we didn't even have emails at all! So given my experience and what I do day-in, day-out, I love that the speed at which technology is developing still astounds me.

The review you are reading was written by a human, not a machine. This fact would no doubt disappoint some of the pioneers of artificial intelligence, who would have thought that by the 21st century a computer would be able to read a book, consider it in the context of other knowledge and express some thoughtful opinions about it. On the other hand, the human who wrote this review was aided in researching and preparing it by telecommunications and computer networks, including the Internet, that owe a big part of their existence and even more of their smooth functioning to theories and concepts that arose from artificial-intelligence research. The enormous, if stealthy, influence of AI bears out many of the wonders foretold 25 years ago in Machines Who Think, Pamela McCorduck s groundbreaking survey of the history and prospects of the field. A novelist at the time (she has since gone on to write and consult widely on the intellectual impact of computing), McCorduck got to the founders of the field while they were still feeling their way into a new science.

Machine learning is ubiquitous in our daily lives. Every time we talk to our smartphones, search for images or ask for restaurant recommendations, we are interacting with machine learning algorithms. They take as input large amounts of raw data, like the entire text of an encyclopedia, or the entire archives of a newspaper, and analyze the information to extract patterns that might not be visible to human analysts. But when these large data sets include social bias, the machines learn that too. If the source documents reflect gender bias – if they more often have the word'doctor' near the word'he' than near'she,' and the word'nurse' more commonly near'she' than'he' – then the algorithm learns those biases too, the researcher explains According to James Zou, Assistant Professor for Biomedical Data Science at Stanford University, machine systems are learning human biases when examples of such are included in the training set.

The literature on judgment aggregation is moving from studying impossibility results regarding aggregation rules towards studying specific judgment aggregation rules. Here we give a structured list of most rules that have been proposed and studied recently in the literature, together with various properties of such rules. We first focus on the majority-preservation property, which generalizes Condorcet-consistency, and identify which of the rules satisfy it. We study the inclusion relationships that hold between the rules. Finally, we consider two forms of unanimity, monotonicity, homogeneity, and reinforcement, and we identify which of the rules satisfy these properties.

Many machine learning tasks can be formulated as Regularized Empirical Risk Minimization (R-ERM), and solved by optimization algorithms such as gradient descent (GD), stochastic gradient descent (SGD), and stochastic variance reduction (SVRG). Conventional analysis on these optimization algorithms focuses on their convergence rates during the training process, however, people in the machine learning community may care more about the generalization performance of the learned model on unseen test data. In this paper, we investigate on this issue, by using stability as a tool. In particular, we decompose the generalization error for R-ERM, and derive its upper bound for both convex and non-convex cases. In convex cases, we prove that the generalization error can be bounded by the convergence rate of the optimization algorithm and the stability of the R-ERM process, both in expectation (in the order of $\mathcal{O}((1/n)+\mathbb{E}\rho(T))$, where $\rho(T)$ is the convergence error and $T$ is the number of iterations) and in high probability (in the order of $\mathcal{O}\left(\frac{\log{1/\delta}}{\sqrt{n}}+\rho(T)\right)$ with probability $1-\delta$). For non-convex cases, we can also obtain a similar expected generalization error bound. Our theorems indicate that 1) along with the training process, the generalization error will decrease for all the optimization algorithms under our investigation; 2) Comparatively speaking, SVRG has better generalization ability than GD and SGD. We have conducted experiments on both convex and non-convex problems, and the experimental results verify our theoretical findings.

Motivated by the need to automate medical information extraction from free-text radiological reports, we present a bi-directional long short-term memory (BiLSTM) neural network architecture for modelling radiological language. The model has been used to address two NLP tasks: medical named-entity recognition (NER) and negation detection. We investigate whether learning several types of word embeddings improves BiLSTM's performance on those tasks. Using a large dataset of chest x-ray reports, we compare the proposed model to a baseline dictionary-based NER system and a negation detection system that leverages the hand-crafted rules of the NegEx algorithm and the grammatical relations obtained from the Stanford Dependency Parser. Compared to these more traditional rule-based systems, we argue that BiLSTM offers a strong alternative for both our tasks.

To date, the instability of prognostic predictors in a sparse high dimensional model, which hinders their clinical adoption, has received little attention. Stable prediction is often overlooked in favour of performance. Yet, stability prevails as key when adopting models in critical areas as healthcare. Our study proposes a stabilization scheme by detecting higher order feature correlations. Using a linear model as basis for prediction, we achieve feature stability by regularising latent correlation in features. Latent higher order correlation among features is modelled using an autoencoder network. Stability is enhanced by combining a recent technique that uses a feature graph, and augmenting external unlabelled data for training the autoencoder network. Our experiments are conducted on a heart failure cohort from an Australian hospital. Stability was measured using Consistency index for feature subsets and signal-to-noise ratio for model parameters. Our methods demonstrated significant improvement in feature stability and model estimation stability when compared to baselines.