The only person in kimono at a recent government meeting on flying cars was Kotaro Chiba, a former online-game executive turned financier of a very specific kind. For Chiba, 44, who wears kimono on special occasions to show his pride in Japanese culture, is gathering money for what he calls the Drone Fund. It invests in unmanned vehicles to survey buildings, make deliveries and take aerial photos for tourist boards; hover scooters; and a pilotless cargo craft that's seeking to make it all the way from Japan to Silicon Valley in one go. Chiba is at the forefront of an industry that's only years away from changing our lives. In five to 10 years, the skies could be alive with drones delivering goods, according to McKinsey & Co.

In Haidilao International Holding Ltd.'s hot pot restaurants, robots are replacing chefs and waiters. Asia's biggest listed restaurant chain by market value is partnering with Panasonic Corp. to open what the two companies say is the world's first eatery with a fully automated kitchen on Oct. 28 in Beijing. At the new Haidilao restaurant, robots will take orders, prepare and deliver raw meat and fresh vegetables to customers to plop into soups prepared at their tables. The automation will lower labor costs and boost efficiency, underpinning Haidilao's plan to expand to as many as 5,000 restaurants worldwide, the companies said. "It could be difficult to expand to that size in terms of personnel, so Haidilao is shifting earlier to an operation that doesn't rely so much on manual labor," said Jun Yamashita, managing director of Ying Hai Holding Pte., the Singapore-based joint venture Haidilao and Panasonic have set up. "That's where Panasonic's technology comes in."

Artificial Intelligence (AI), Machine Learning, Deep Learning etc. are not new terms. AI has been around for a decade now and Gartner predicts it will be a top five investment priority by 2020. AI's primary way of personalizing each stage of consumer journey has overwhelmed the world and has opened new doors for marketing and advertising. Siri, Alexa, Google assistant are making it easy for people to search anything with just a voice command or with the press of a button. Recommender systems is now things of the past and a primary platform for online shopping sites to earn more revenue.

Indian Prime Minister, Narendra Modi has recently expressed positive sentiment regarding the use of emerging technologies such as Blockchain and Artificial Intelligence as being drivers of economic growth. Continued automation and digitisation has led many people to grow fearful of losing their jobs to AI and Machine Learning. However, PM Modi claims that these technologies would not eliminate the need for human employment but would change the nature of the work done. Here we take a look at the various ways in which the World Economic Forum and NITI Aayog plan to use emerging tech for increased employment. Speaking at the launch of the fourth centre for the World Economic Forum in Maharashtra, PM Modi addressed the fear of critics that automation would lead to a decrease in employment.

It was just reported that Haidilao, a 5,000-restaurant chain in Asia, intends to save on labor costs by replacing chefs and waiters with robots. To "lower labor costs and boost efficiency." Restaurants succeed by serving great, nutritious fresh food in a nice atmosphere staffed by great people. How many readers have stopped by a Horn & Hardart automat lately? Horn & Hardart helped pioneer the automat concept as early as 1902, but where are all those automats now? That's why it's encouraging that the companies that are taking an early lead with artificial intelligence don't see it as just labor-saving technology -- rather, they see it as a way to grow, to better serve customers and provide a superior experience.

Karthik Sundaram, Program Manager-The Industrial Internet of Things, Frost & Sullivan – an excerpt from SPS IPC Drives 2018 presentation to be delivered 28th of November 2018 at 2.00-2.30 Situated in a mountain village of Japan is FANUC's widely reported lights out factory. This one of a kind, unmanned factory works autonomously 24/7 and is well known for robots that can assemble, test, and monitor themselves. A few decades ago, such a scenario would have existed only in the pages of Isaac Asimov's science fiction. Today, the FANUC use case is a proof of the dawn of cognitive factories and how far artificial intelligence (AI) has been able to penetrate into the walls of these factories.

The Expectation-Maximization (EM) algorithm for mixture models often results in slow or invalid convergence. The popular convergence proof affirms that the likelihood increases with Q; Q is increasing in the M -step and non-decreasing in the E-step. The author found that (1) Q may and should decrease in some E-steps; (2) The Shannon channel from the E-step is improper and hence the expectation is improper. The author proposed the CM-EM algorithm (CM means Channel's Matching), which adds a step to optimize the mixture ratios for the proper Shannon channel and maximizes G, average log-normalized-likelihood, in the M-step. Neal and Hinton's Maximization-Maximization (MM) algorithm use F instead of Q to speed the convergence. Maximizing G is similar to maximizing F. The new convergence proof is similar to Beal's proof with the variational method. It first proves that the minimum relative entropy equals the minimum R-G (R is mutual information), then uses variational and iterative methods that Shannon et al. use for rate-distortion functions to prove the global convergence. Some examples show that Q and F should and may decrease in some E-steps. For the same example, the EM, MM, and CM-EM algorithms need about 36, 18, and 9 iterations respectively.

Ensemble methods are among the state-of-the-art predictive modeling approaches. Applied to modern big data, these methods often require a large number of sub-learners, where the complexity of each learner typically grows with the size of the dataset. This phenomenon results in an increasing demand for storage space, which may be very costly. This problem mostly manifests in a subscriber based environment, where a user-specific ensemble needs to be stored on a personal device with strict storage limitations (such as a cellular device). In this work we introduce a novel method for lossless compression of tree-based ensemble methods, focusing on random forests. Our suggested method is based on probabilistic modeling of the ensemble's trees, followed by model clustering via Bregman divergence. This allows us to find a minimal set of models that provides an accurate description of the trees, and at the same time is small enough to store and maintain. Our compression scheme demonstrates high compression rates on a variety of modern datasets. Importantly, our scheme enables predictions from the compressed format and a perfect reconstruction of the original ensemble. In addition, we introduce a theoretically sound lossy compression scheme, which allows us to control the trade-off between the distortion and the coding rate.

Online learning makes sequence of decisions with partial data arrival where next movement of data is unknown. In this paper, we have presented a new idea as multiple times weight updating that update the weight iteratively for same instance. The proposed technique analyzed with popular algorithms from literature and experimented using established tool. The results indicates that mistake rate reduces to zero or close to zero for various datasets and algorithms. The overhead running cost is not too expensive and achieving mistake rate close to zero further strengthen the proposed technique. The proposed technique could be helpful to meet real life challenges.

In this article, we show that solving the system of linear equations by manipulating the kernel and the range space is equivalent to solving the problem of least squares error approximation. This establishes the ground for a gradient-free learning search when the system can be expressed in the form of a linear matrix equation. When the nonlinear activation function is invertible, the learning problem of a fully-connected multilayer feedforward neural network can be easily adapted for this novel learning framework. By a series of kernel and range space manipulations, it turns out that such a network learning boils down to solving a set of cross-coupling equations. By having the weights randomly initialized, the equations can be decoupled and the network solution shows relatively good learning capability for real world data sets of small to moderate dimensions. Based on the structural information of the matrix equation, the network representation is found to be dependent on the number of data samples and the output dimension.