Spectral dimensionality reduction is frequently used to identify low-dimensional structure in high-dimensional data. However, learning manifolds, especially from the streaming data, is computationally and memory expensive. In this paper, we argue that a stable manifold can be learned using only a fraction of the stream, and the remaining stream can be mapped to the manifold in a significantly less costly manner. Identifying the transition point at which the manifold is stable is the key step. We present error metrics that allow us to identify the transition point for a given stream by quantitatively assessing the quality of a manifold learned using Isomap. We further propose an efficient mapping algorithm, called S-Isomap, that can be used to map new samples onto the stable manifold. We describe experiments on a variety of data sets that show that the proposed approach is computationally efficient without sacrificing accuracy.

Waymo CEO John Krafcik speaks at a press conference at the 2017 North American International Auto Show in Detroit on Jan. 8, 2017. Waymo, the company born from Alphabet's Google Self-Driving Car research project, is designing and building all the sensors, radar and computers used in its automated test vehicles, along with the artificial intelligence programs that control everything. Yet to make its technology affordable for commercial use, it anticipates a manufacturing alliance as it looks ahead to mass-scale production of components, according to Chief Executive Officer John Krafcik. Waymo this week at the North American International Auto Show in Detroit revealed that the latest generation of its hardware and software is being used on Chrysler Pacifica minivans that begin road tests this month. A total of 100 of the vans are getting radar, sensors, cameras and laser Lidar units for 360-degree, high-definition images of a vehicle's surrounding, all made by Waymo.

Edited by Michael F. Goodchild, University of California, Santa Barbara, CA, and approved November 22, 2016 (received for review July 20, 2016) Ride-sharing services can provide not only a very personalized mobility experience but also ensure efficiency and sustainability via large-scale ride pooling. Large-scale ride-sharing requires mathematical models and algorithms that can match large groups of riders to a fleet of shared vehicles in real time, a task not fully addressed by current solutions. We present a highly scalable anytime optimal algorithm and experimentally validate its performance using New York City taxi data and a shared vehicle fleet with passenger capacities of up to ten. Our results show that 2,000 vehicles (15% of the taxi fleet) of capacity 10 or 3,000 of capacity 4 can serve 98% of the demand within a mean waiting time of 2.8 min and mean trip delay of 3.5 min. Ride-sharing services are transforming urban mobility by providing timely and convenient transportation to anybody, anywhere, and anytime. Current mathematical models, however, do not fully address the potential of ride-sharing. Recently, a large-scale study highlighted some of the benefits of car pooling but was limited to static routes with two riders per vehicle (optimally) or three (with heuristics).

Global technology major Google, which is primarily accredited for developing the search engine and transforming mobiles into complex devices, is banking on the concept of artificial intelligence (AI) and machine learning to drive its future growth. Already implementing complex algorithms in its search engine, the company is now integrating these concepts and coming up with solutions which can be used and implemented in daily life like Google Maps. The company's global chief executive officer (CEO), Sundar Pichai said a solution which can track blindness in early stages so that it can be cured before it takes an acute shape is being worked upon. Traces of such machine learning techniques can be found in Google Translate which Pichai claimed, had become more powerful than it was 10 years before. Pichai said India had the potential to emerge as a technology leader in the forthcoming times, and solutions which are made for India specifically can be implemented globally as well.

Here I offer a few off-the-beaten-path interesting problems that you won't find in textbooks, data science camps, or in college classes. These problems range from applied maths, to statistics and computer science, and are aimed at getting the novice interested in a few core subjects that most data scientists master. The problems are described in simple English and don't require math / stats / probability knowledge beyond high school level. My goal is to attract people interested in data science, but who are somewhat concerned by the depth and volume of (in my opinion) unnecessary mathematics included in many curricula. I believe that successful data science can be engineered and deployed by scientists coming from other disciplines, who do not necessarily have a deep analytical background yet are familiar with data.

In 2015, several of the world's top poker players faced down a supercomputer-powered artificial intelligence named Claudico during a grueling 80,000 hands of no-limit Texas Hold'em. Beginning tomorrow, a rematch of humans versus AI will test whether humanity can hold its own against an even more capable challenger. The human margin of victory from the past event was not large enough to statistically prove whether humans or the Claudico AI were really the better poker players. This year's rematch features four human poker pros playing for a prize pot of $200,000 against an AI called Libratus in the "Brains Vs. Artificial Intelligence: Upping the Ante" event being held at the Rivers Casino in Pittsburgh starting on 11 January.

From a nation defined by cricket, steam engines and horses to a celebrity-obsessed, football-mad country, Britain has undergone huge cultural changes between 1800 and 1950. The history of the country is well-documented in books, but a new analysis has used a different source, newspapers, to pinpoint exactly when these shifts in culture took place. A group of computer scientists have developed artificial intelligence software to analyse articles from 120 British newspapers between 1800 and 1950, and revealed key turning points in our culture. In 2011, a huge analysis of five million books taken from over 200 years provided the first clues to what insights can be gained from analysing words on such a huge scale. But the project was criticised for only counting words, and ignored the context they were based in.

Self-driving cars will perform rationally. For example: stop when someone is in their way. Research suggests humans will take advantage, and step into an intersection when they know they shouldn't.

IBM Watson Health's Watson for Genomics platform is to help inform cancer treatment decisions based on gene sequencing data from company Illumina. The artificial intelligence software will by applied to data from Illumina's TruSight Tumor 170 platform – a next generation sequencing gene panel that provides data related to 170 genes associated with common solid tumours. Watson will analyse the data alongside medical literature, clinical trial data and professional guidelines to help produce informed insights on how cancers could be treated. The entire process from raw genetic data to treatment insights will be completed in a matter of minutes – much quicker than the usual week or so it takes scientists to produce a similar report. "This partnership lays the groundwork for more systematic study of the impact of genomics in oncology," said Deborah DiSanzo, general manager, IBM Watson Health.

It was a disappointing holiday season for French drone maker Parrot. The company announced Monday that it will lay off about 290 employees -- or more than a third of the employees currently working on drone-related projects -- after it missed fourth quarter sales targets by about 15 percent. Parrot's drone line has always straddled the line between executive toys and high-end gadgetry, and according to the company's announcement, the first order of business in 2017 will be to refocus its product offerings to concentrate on the most profitable areas. Parrot cites extremely low margins in the consumer drone space as the main reason for the layoffs, so we won't be surprised if we see fewer one-off jumping, flying and seafaring minidrones as the company continues to grow up with more prosumer-level devices like the Bebop 2 or the fixed-wing Disco in the future. As Recode notes, Parrot's main competitor DJI has been dominating the consumer drone space by slashing prices and cutting into profit margins, which it can pull off because it also owns the manufacturing facilities in Shenzhen.