When the Samsung Galaxy S8 was released in April, users were disappointed to find that their device was missing Bixby, the firm's highly-anticipated digital assistant. And now, it appears that the launch of Bixby may have been struck by further delays. A new report suggests that Samsung's digital assistant is'struggling to comprehend English' forcing developers to continue working on it. It appears that the launch of Bixby may have been struck by delays. A new report suggests that Samsung's digital assistant is'struggling to comprehend English' forcing developers to continue working on it According to Samsung, Bixby is'a completely new way to use your Galaxy S8 or S8 .' Users can use voice, text, or touch to say what they need, since it understands all three.

We study the combinatorial pure exploration problem Best-Set in stochastic multi-armed bandits. In a Best-Set instance, we are given $n$ arms with unknown reward distributions, as well as a family $\mathcal{F}$ of feasible subsets over the arms. Our goal is to identify the feasible subset in $\mathcal{F}$ with the maximum total mean using as few samples as possible. The problem generalizes the classical best arm identification problem and the top-$k$ arm identification problem, both of which have attracted significant attention in recent years. We provide a novel instance-wise lower bound for the sample complexity of the problem, as well as a nontrivial sampling algorithm, matching the lower bound up to a factor of $\ln|\mathcal{F}|$. For an important class of combinatorial families, we also provide polynomial time implementation of the sampling algorithm, using the equivalence of separation and optimization for convex program, and approximate Pareto curves in multi-objective optimization. We also show that the $\ln|\mathcal{F}|$ factor is inevitable in general through a nontrivial lower bound construction. Our results significantly improve several previous results for several important combinatorial constraints, and provide a tighter understanding of the general Best-Set problem. We further introduce an even more general problem, formulated in geometric terms. We are given $n$ Gaussian arms with unknown means and unit variance. Consider the $n$-dimensional Euclidean space $\mathbb{R}^n$, and a collection $\mathcal{O}$ of disjoint subsets. Our goal is to determine the subset in $\mathcal{O}$ that contains the $n$-dimensional vector of the means. The problem generalizes most pure exploration bandit problems studied in the literature. We provide the first nearly optimal sample complexity upper and lower bounds for the problem.

In machine learning ensemble methods have demonstrated high accuracy for the variety of problems in different areas. The most known algorithms intensively used in practice are random forests and gradient boosting. In this paper we present InfiniteBoost -- a novel algorithm, which combines the best properties of these two approaches. The algorithm constructs the ensemble of trees for which two properties hold: trees of the ensemble incorporate the mistakes done by others; at the same time the ensemble could contain the infinite number of trees without the over-fitting effect. The proposed algorithm is evaluated on the regression, classification, and ranking tasks using large scale, publicly available datasets.

Over the last years, the world has witnessed the emergence of data sets of an unprecedented size across different scientific disciplines. The large volume of such data sets presents new challenges as gathering, storing, and analyzing them becomes expensive. In the context of millions or even billions of data points, existing proven algorithms "suddenly" become computationally infeasible while data sets may not fit on single machines anymore but must be stored on clusters of machines. As a consequence, new algorithms are required to scale to this massive data setting. While one could focus on single machine learning problems and come up with endless new algorithms, we focus on a more general approach: we investigate coresets -- succinct, small summaries of large data sets -- so that solutions found on the summary are provably competitive with solution found on the full data set.

Science fiction author and astrophysicist David Brin believes humans have a range of options to consider to prevent artificial intelligence from ruling over people the way kings and foreign invaders. Asimov's Three Law of Robotics and regulation are key, but so is being wary of manipulation. "The first robotic empathy crisis is going to happen very soon," Brin said. "Within three to five years we will have entities either in the physical world or online who demand human empathy, who claim to be fully intelligent, and claim to be enslaved beings, enslaved artificial intelligences, and who sob and demand their rights." Thousands upon thousands of protestors will be in the streets demanding rights for AI, Brin predicts, and those who aren't immediately convinced will be analyzed.

It's convenient when Facebook can tag your friends in photos for you, and it's fun when Snapchat can apply a filter to your face. Both are examples of algorithms that have been trained to recognize eyes, noses, and mouths with consistent accuracy. When these programs are wrong--like when Facebook mistakes you for your sibling or even your mom--it's hardly a problem. In other situations, though, we give artificial intelligence much more responsibility, with larger consequences when it inevitably backfires. Ben Shneiderman, a computer scientist from the University of Maryland, thinks the risks are big enough that it's time to for the government to get involved. In a lecture on May 30 to the Alan Turing Institute in London, he called for a "National Algorithm Safety Board," similar to the US's National Transportation Safety Board for vehicles, which would provide both ongoing and retroactive oversight for high-stakes algorithms.

Unless you've recently graduated from one of the new Data Science courses that have been popping up online and in various universities around the world, then becoming a Data Scientist was most likely slightly accidental and was more about the journey than the destination. I started out as a physicist and had a strong mathematical grounding, but I had a passion for medicine. After completing my bachelor's degree I took a master's degree in medical physics. This is where I gained an appreciation for the importance of image analysis and the role that data plays in medicine. I created a virtual model of a human torso by segmenting images from the Visible Human Project.

It may have been a while since we last we came across a Here 3D mapping vehicle, but that's not to say the company hasn't been using its cars lately. In fact, the Here True collection vehicle is now in its third revision, and I got to hop on one -- based on a Volkswagen Golf Variant 280 TSI Highline -- during Computex. The ride features much faster D-GPS tracking that no longer requires a half-hour calibration (to reach an accuracy of under one meter), along with a Velodyne LiDAR with an accuracy of better than 2cm (within a range of up to 70 meters) and four 16.2-megapixel MARS panoramic cameras. This set of gear is almost identical to what we've seen before, so the real highlight this time is the updated backend to support high-definition mapping. For HD mapping, Here's fleet is currently involved in over 20 projects worldwide and covering over 48,200 km per week, with each collection vehicle contributing around 500 km or over 1TB worth of data.

Together, the last two responses add up to 71 percent of respondents, and Accenture reports that insurers are investing in AI in several areas of the business, including distribution, claims and underwriting. They are looking to empower agents, brokers and employees to enhance the customer experience with automated personalized services, faster claims handling and individual risk-based underwriting processes, Accenture said in a statement. Separately, Carrier Management interviewed representatives of four global insurance groups--XL Catlin, Allianz, QBE and Zurich--who described some of the AI initiatives already underway at their firms. They range from condensing lengthy engineering reports for swifter underwriting, reassigning some of the claims administrative services handled by offshore humans to robots, interpreting crop risk information delivered by drones, and deciphering communications from customers with heavy accents using natural language processing. Also detailed is Zurich's use of a multilingual natural language processing product called Cogito from Expert Systems that mines complicated, voluminous claims data to rapidly provide more refined information to claims adjusters for faster decision-making.

Big questions about humanity's future also merit a side question over humanity's future alongside machines.How long before AI outperforms humans in various tasks? A bracing new study presents responses from professionals working in that field. Evidence from AI Experts," is by Katja Grace, John Salvatier, Allan Dafoe, Baobao Zhang and Owain Evans, and it is on arXiv. Actually, this paper reveals not if this will happen but when, asking how many things will AI be able to do even better than we can? "Researchers believe there is a 50% chance of AI outperforming humans in all tasks in 45 years and of automating all human jobs in 120 years." Interestingly, Asian respondents expected the dates much sooner than did North Americans.