Businesses are investing heavily in digitalisation and are increasingly using computer-controlled processes in their various departments. They are also investing in artificial intelligence (AI) to an ever great extent. With AI, technology makes autonomous decisions (in some cases), so human intervention is no longer necessary. HR professionals who use AI mainly do so to screen CVs and to send messages to specific target groups automatically via social media, and for staff training and development, for example. According to a survey of 1,000 HR professionals in Belgium, the Netherlands, France, Germany and the UK, conducted by iVOX on behalf of SD Worx, six in 10 European HR professionals believe AI will lead to job losses in their sector by 2025.

What does automation mean for the UK manufacturing, supply chain and logistics sector? In July 2017, leaders in manufacturing, supply chain and logistics from some of the UK's top businesses joined Ben Gilbert & Rachel McGowan-Kemp to discuss human and machine collaboration – both today and what automation means for the future. In our new discussion document we address the impact on the industry of how humans and machines are collaborating. The paper outlines the thoughts, opinions and trends voiced by industry leaders in supply chain, manufacturing and logistics at the roundtable dinner. Over the last twenty years, technology has rapidly altered the nature of industry.

The government's'Nudge Unit' is experimenting with using machine learning algorithms to rate how well schools and doctors' surgeries are performing. For the last year, The Behavioural Insights Team (BIT) has been trialling machine learning models that can crunch through publicly available data to help automate some of the decisions made by bodies such as Ofsted, which inspects schools, and the Care Quality Commission, which regulates health and social care in England. Michael Sanders, head of research at the BIT says it is working with Ofsted to put the technology into use during 2018. "We're working with them to feed into variations on our model and to improve it using additional data that they have that isn't public," he says. The school-evaluating algorithm pulls together data from a large number of sources to decide whether a school is potentially performing inadequately. It is said the system can help to identify more schools that are inadequate, when compared to random inspections.

I don't have to open the doors of AImotive's white 2015 Prius to see that it's not your average car. This particular Prius has been christened El Capitan, the name written below the rear doors, and two small cameras are mounted on top of the car. Inside is where things really get interesting, though. The trunk holds a computer the size of a microwave, and a large monitor covers the passenger glove compartment and dashboard. The center console has three switches labeled "Allowed," "Error," and "Active."

A new generation of fintech companies is building digital solutions to transform the investment and asset management industry. Often referred to as "wealthtech" companies, these use artificial intelligence (AI) and big data to offer advice, and provide low-cost alternatives with micro-investment platforms and trading solutions based on social networks. Wealthtech reached a record of 74 financing agreements worth US$657 million in 2016 and has been growing steadily for the past five years, according to data from CB Insights. And while robo-advisors (automated services that use machine-learning algorithms to offer personalized users advice) are by far the best-known product in the field, there are many new ideas emerging that seek to improve the investment industry through technology. Robo-retirements, for instance, are robo-advisors that specialize in managing retirement savings.

At the end of every year, tech PR firms circulate and hawk the prognostications of their client companies' executives on what the next year will bring in the world of data and analytics. There are almost always contradictions to be found on certain points and suspicious unanimity on others. And because the predictions tend to function as self-serving marketing messages, sometimes they can sound more like taglines than substantive forecasts. It's always fun to read and sort out these predictions. That may sound a bit snarky but -- I gotta say -- even if it's a lot of work, it's always fun to read and sort out these predictions.

Artificial intelligence (AI) is designed to put us out of the picture. Still, we shouldn't fret since a recently published study discovered how humans can be on top of things in systems that rely on AI. The study, carried out by researchers at the Ecole Polytechnique Fédérale de Lausanne (EPFL), explained that while AI will always find ways to bypass human intervention and build an independent solution, operators must look for ways to keep themselves above the machines and prevent them from circumventing human command. The solution that the researchers found was to change the rules midstream: To borrow a psychological term, instead of punishing AI for learning the process and gaining independence, operators opined that they be one step ahead of the machines and keep leading them by moving the proverbial carrot. In AI, machines are programmed to learn from their tasks -- the do an activity, observe what happens, adapt their behavior, and apply it to the next action.

While Siri and Alexa rely on a plethora of components and cloud servers to work out what people are saying, our brain can do it with a tiny area, researchers have revealed. We are able to distinguish Donald Trump's speech from the voice of our mother due to a section of the right posterior temporal lobe, one of the four major lobes in a mammalian brain, new research from a German university has found. Scientists have long disagreed on the exact area responsible for voice recognition, but they now believe it all takes place in a tiny area is known as the superior temporal gyrus (STG). These findings were supported by a previous study at MPI CBS, whereby, a phenomenon commonly known as'voice blindness', phonagnosia, the inability to recognize voices, was investigated. The two involved study participants - the only ones known in Germany to be affected by phonagnosia - were not able to identify people by voice, even their own mum or child.

Clustering partitions a dataset such that observations placed together in a group are similar but different from those in other groups. Hierarchical and $K$-means clustering are two approaches but have different strengths and weaknesses. For instance, hierarchical clustering identifies groups in a tree-like structure but suffers from computational complexity in large datasets while $K$-means clustering is efficient but designed to identify homogeneous spherically-shaped clusters. We present a hybrid non-parametric clustering approach that amalgamates the two methods to identify general-shaped clusters and that can be applied to larger datasets. Specifically, we first partition the dataset into spherical groups using $K$-means. We next merge these groups using hierarchical methods with a data-driven distance measure as a stopping criterion. Our proposal has the potential to reveal groups with general shapes and structure in a dataset. We demonstrate good performance on several simulated and real datasets.

We derive a novel variational expectation maximization approach based on truncated variational distributions. Truncated distributions are proportional to exact posteriors within a subset of a discrete state space and equal zero otherwise. The novel variational approach is realized by first generalizing the standard variational EM framework to include variational distributions with exact (`hard') zeros. A fully variational treatment of truncated distributions then allows for deriving novel and mathematically grounded results, which in turn can be used to formulate novel efficient algorithms to optimize the parameters of probabilistic generative models. We find the free energies which correspond to truncated distributions to be given by concise and efficiently computable expressions, while update equations for model parameters (M-steps) remain in their standard form. Furthermore, we obtain generic expressions for expectation values w.r.t. truncated distributions. Based on these observations, we show how efficient and easily applicable meta-algorithms can be formulated that guarantee a monotonic increase of the free energy. Example applications of the here derived framework provide novel theoretical results and learning procedures for latent variable models as well as mixture models including procedures to tightly couple sampling and variational optimization approaches. Furthermore, by considering a special case of truncated variational distributions, we can cleanly and fully embed the well-known `hard EM' approaches into the variational EM framework, and we show that `hard EM' (for models with discrete latents) provably optimizes a lower free energy bound of the data log-likelihood.