New research from Accenture shows that £218 million has been invested in insurance technology startups in the first half of 2017 – up from £7.8 million the previous year. This represents an increase of over 2500% in year-over-year figures. Investment in insurance technology startups, or insurtech, have been bolstered by some recent, extremely large investments including the £180 million invested in Gryphon, an insurtech startup focused on critical illness and income protection. Just last week, Coya, a Berlin-based digital insurance provider founded in 2016 managed to raise $10 million (approx £7.7 million) from a variety of investors from Silicon Valley and Europe. Various new technologies are being applied to the insurance industry, which has historically been largely administrative and paper intensive.

Artificial intelligence (AI) may be the next big thing for communications service providers (CSPs), but it's not clear yet exactly how they will use it or where it will have the biggest impact on their business. "Let's break it down a bit – it can be misleading," Telefónica Global Group's CIO, Phil Jordan, told attendees at the Executive Summit during TM Forum Live!. "We see clear use cases and value in cognitive and machine learning. Any decision we take in a systemic way, I have asked for a plan for when and where does that become a machine-learned activity? "It's the next transformation wave that is going to hit all of us – converting decision-making into something that isn't static rule-based," he adds. "I don't think that's a technology problem – it's here or it's coming." But is Telefónica making extensive use of AI today? "We made no use of it in the transformation," Jordan emphasized in discussing the company's massive digital transformation. "AI isn't a magic trick," he says, and it won't be useful unless operators transform their existing IT systems first. Indeed, that's the message we've been hearing from many of our members: AI is promising but it isn't reality – yet. In November we will publish an extensive Trend Analysis Report on AI and machine learning, analyzing the results of our surveys of CSPs and suppliers (choose the right one for you). Take the survey and you'll be entered into a draw for a $250 Amazon gift voucher. Certainly, new virtualized network functionality and new operational and business support systems are needed to take advantage of AI and machine learning (a form of AI), in customer facing applications such as virtual agents and chatbots and for end-to-end network and service management. "You have to teach it; you have to give the machine context all the time," Jordan explains. "You have to have a business that is ready and able to understand outcomes and go back and feed it into machine learning.

A sheen is starting to appear on Rocky Blumhagen's forehead, just below his gray hair. He's marching in place in a starkly lit room decked out with two large flatscreens. On both of the TVs, a volcano lets off steam through wide cracks glowing with lava, their roar muffling the Andean percussion and flutes on the soundtrack. Rocky reaches out his left hand, as if to grasp a coin from midair, and one of them disappears with a brrring. "I don't know if I can do it," he says to a guy named Josh sitting nearby in a felt-covered lounge chair. He looks up from his iPad, watching Rocky, age 66, grab, jog, kick, and reach his way through the videogame. "Keep it up," Josh says as the heart monitor in the corner of the screen reads 129.

Since its messaging app debuted in June 2011, Line Corp. has shaken up the online communications landscape in Japan and morphed into a player in smartphone communications infrastructure. So Line is planting the seeds of success for what it thinks will be the next big thing: voice-based "AI agents." While this artificial-intelligence quest will pit the smaller Line against IT powerhouses Google, Apple and Amazon, among others, Line CEO Takeshi Idezawa likes his chances. "We are taking on a new challenge because we believe we have the assets to win the battle," Idezawa told The Japan Times in a recent interview. This is quite a change for a firm that owes its success to a prescient bet on smartphones less than a decade ago.

In this course we are going to look at advanced NLP. Previously, you learned about some of the basics, like how many NLP problems are just regular machine learning and data science problems in disguise, and simple, practical methods like bag-of-words and term-document matrices. These allowed us to do some pretty cool things, like detect spam emails, write poetry, spin articles, and group together similar words. In this course I'm going to show you how to do even more awesome things. We'll learn not just 1, but 4 new architectures in this course.

This course is the next logical step in my deep learning, data science, and machine learning series. I've done a lot of courses about deep learning, and I just released a course about unsupervised learning, where I talked about clustering and density estimation. So what do you get when you put these 2 together? In these course we'll start with some very basic stuff - principal components analysis (PCA), and a popular nonlinear dimensionality reduction technique known as t-SNE (t-distributed stochastic neighbor embedding). Next, we'll look at a special type of unsupervised neural network called the autoencoder.

Earthquakes are, by their nature, unpredictable. Although geologists understand why and how the tremors occur, forecasting them more than a few minutes ahead is very difficult. A team of scientists believes that machine learning could help solve this problem one day. A paper published Wednesday in the Geophysical Research Letters describes a method that relies on listening for acoustic signals from a laboratory simulation of failing fault lines. Stress is applied to two heavy steel blocks, causing them to slip and slide over one another like tectonic plates during an earthquake.

It may not have been "The Matrix"[1], but the machines look like they are finally poised to take our jobs. Machines powered by artificial intelligence and machine learning process data faster, aren't hindered by stupid human biases, don't waste time with gossip on social media and don't demand raises or more days off. Figure 1: Is Artificial Intelligence Putting Humans Out of Work? While there is a high probability that machine learning and artificial intelligence will play an important role in whatever job you hold in the future, there is one way to "future-proof" your career…embrace the power of design thinking. I have written about design thinking before, but I want to use this blog to provide more specifics about what it is about design thinking that can help you to harness the power of machine learning…instead of machine learning (and The Matrix) harnessing you. Design thinking is defined as human-centric design that builds upon the deep understanding of our users (e.g., their tendencies, propensities, inclinations, behaviors) to generate ideas, build prototypes, share what you've made, embrace the art of failure (i.e., fail fast but learn faster) and eventually put your innovative solution out into the world.

Detecting and segmenting salient objects in natural scenes, often referred to as salient object detection, has attracted a lot of interest in computer vision. While many models have been proposed and several applications have emerged, yet a deep understanding of achievements and issues is lacking. We aim to provide a comprehensive review of the recent progress in salient object detection and situate this field among other closely related areas such as generic scene segmentation, object proposal generation, and saliency for fixation prediction. Covering 228 publications, we survey i) roots, key concepts, and tasks, ii) core techniques and main modeling trends, and iii) datasets and evaluation metrics in salient object detection. We also discuss open problems such as evaluation metrics and dataset bias in model performance and suggest future research directions.

In this paper, we propose a probabilistic parsing model that defines a proper conditional probability distribution over non-projective dependency trees for a given sentence, using neural representations as inputs. The neural network architecture is based on bidirectional LSTM-CNNs, which automatically benefits from both word-and character-level representations, by using a combination of bidirectional LSTMs and CNNs. On top of the neural network, we introduce a probabilistic structured layer, defining a conditional log-linear model over non-projective trees. By exploiting Kirchhoff's Matrix-Tree Theorem (Tutte, 1984), the partition functions and marginals can be computed efficiently, leading to a straightforward end-to-end model training procedure via back-propagation. We evaluate our model on 17 different datasets, across 14 different languages. Our parser achieves state-of-the-art parsing performance on nine datasets.