Self-organizing maps (SOM) and neural networks of learning vector quantization (LVQ) have seen extensive use for solving different problems in Data Mining domain (clustering, classification, fault detection and compression of information etc.). This type of neural networks was proposed by T. Kohonen [1, 2] and represents, in fact, a single-layer feedforward architecture, which provides an operator for mapping of input space into the output space. Operation-wise SOM and LVQ are quite similar to each neuron is fed input signal (sample) producing output, which is used during competition stage to determine winning neuron - usually the one with maximum output signal value. Vector of synaptic weights for winning neuron is the one closest to the input sample in terms of the metric chosen (which is Euclidian metric in most cases). Next is neurons adjustment phase.

In this paper, Kernel PCA is reinterpreted as the solution to a convex optimization problem. Actually, there is a constrained convex problem for each principal component, so that the constraints guarantee that the principal component is indeed a solution, and not a mere saddle point. Although these insights do not imply any algorithmic improvement, they can be used to further understand the method, formulate possible extensions and properly address them. As an example, a new convex optimization problem for semi-supervised classification is proposed, which seems particularly well-suited whenever the number of known labels is small. Our formulation resembles a Least Squares SVM problem with a regularization parameter multiplied by a negative sign, combined with a variational principle for Kernel PCA. Our primal optimization principle for semi-supervised learning is solved in terms of the Lagrange multipliers. Numerical experiments in several classification tasks illustrate the performance of the proposed model in problems with only a few labeled data.

Additive utility function models are widely used in multiple criteria decision analysis. In such models, a numerical value is associated to each alternative involved in the decision problem. It is computed by aggregating the scores of the alternative on the different criteria of the decision problem. The score of an alternative is determined by a marginal value function that evolves monotonically as a function of the performance of the alternative on this criterion. Determining the shape of the marginals is not easy for a decision maker. It is easier for him/her to make statements such as "alternativea is preferred tob". In order to help the decision maker, UTA disaggregation procedures use linear programming to approximate the marginals by piecewise linear functions based only on such statements. In this paper, we propose to infer polynomials and splines instead of piecewise linear functions for the marginals. In this aim, we use semidefinite programming instead of linear programming. We illustrate this new elicitation method and present some experimental results. Introduction The theory of value functions aims at assigning a number to each alternative in such a way that the decision maker's preference order on the alternatives is the same as the order on the numbers associated with the alternatives. The number or value associated to an alternative is a monotone function of its evaluations on the various relevant criteria. For preferences satisfying some additional properties (includingpreferential independence), the value of an alternative can be obtained as the sum of marginal value functions each depending only on a single criterion [20, Chapter 6]. These functions usually are monotone, i.e., marginal value functions either increase or decrease with the assessment of the alternative on the associated criterion. Many questioning protocols have been proposed aiming to elicit an additive value function [20, 9] through interactions with the decision maker (DM). These direct elicitation methods are time-consuming and require a substantial cognitive effort from the DM. Therefore, in certain cases, an indirect approach may prove fruitful. The latter consists inlearning an additive value model (or a set of such models) from a set of declared or observed preferences. Learning approaches have been proposed not only for inferring an additive value function that is used to rank all other alternatives.

Israeli startup Faception made headlines this year by claiming it could predict how likely people are to be terrorists, pedophiles, and more by analyzing faces with deep learning. Experts and research in the field, however, suggest that it is more fantasy than reality. Faception assigns ratings after training artificial intelligence on faces of terrorists, pedophiles, Mensa members, professional poker players, and more. Through deep learning--that emerging technique found in everything from Alpha Go to Siri to Netflix--the AI can supposedly predict how likely a new face is to belong to any given group. While this may sound believable, there's no evidence that face-based personality predictions are more than a tiny bit accurate.

A rainbow flag flies with the Union flag above British Cabinet Offices. Thousands of gay and bisexual men convicted under Britain's now-defunct sexual offense laws will be posthumously pardoned. The Ministry of Justice announced the proposed amendment Thursday that would posthumously pardon thousands convicted under those outdated laws. The so-called "Turing's Law" would also allow those who are living to apply to have their names removed from criminal records. Lord John Sharkey, the man behind the amendment, called the development "momentous" and said that of the 65,000 men convicted under the laws, 15,000 are still alive, BBC reported.

John Bassett, a British spy who worked for the agency GCHQ for nearly two decades, has told Daily Express that the U.S. was considering plans to employ thousands of robots by 2025. At a meeting with police and counter-terrorism officials in London, he said: "At some point around 2025 or thereabouts the U.S. army will actually have more combat robots than it will have human soldiers. Many of those combat robots are trucks that can drive themselves, and they will get better at not falling off cliffs. But some of them are rather more exciting than trucks. So we will see in the West combat robots outnumber human soldiers."

The first computers cost millions of dollars and were locked inside rooms equipped with special electrical circuits and air conditioning. The only people who could use them had been trained to write programs in that specific computer's language. Today, gesture-based interactions, using multitouch pads and touchscreens, and exploration of virtual 3D spaces allow us to interact with digital devices in ways very similar to how we interact with physical objects. Multitouch pads and touchscreens recognize movements of fingers on a surface, while devices such as the Wii and Kinect recognize movements of arms and legs. Frances Van Scoy says this is bringing us closer to towards'computing at the speed of thought' A professor at West Virginia University believes her research is helping to move us toward what might be called'computing at the speed of thought.' Frances Van Scoy says low-cost open-source projects such as OpenBCI allow people to assemble their own neuroheadsets that capture brain activity noninvasively.

Google has made minimal forays into real-world retail shops thus far. There's a good reason for that: the company has long been more focused on software than hardware. That's slowly changing over time, but Google went all-in on its own hardware brand when it announced the new Pixel smartphones, Google Home, Daydream VR headset and Google WiFi router earlier this month. For most consumers, buying hardware sight-unseen is still a tough proposition, so Google is finally making it easier for consumers to check out all its new gadgets -- in New York City, at least. The company's pop-up retail location opened its doors this morning, and while it wasn't exactly an iPhone-level stampede, there were a couple dozen people waiting to get in when it opened.

The future of artificial intelligence (AI) taking human jobs as predicted by the White House last week is far less extreme than what English theoretical physicist Stephen Hawking is expecting to happen. "In short, the rise of powerful AI will be either the best or the worst thing ever to happen to humanity," he told an audience at the launch of the new Cambridge University's Leverhulme Center for the Future of Intelligence (CFI) on Wednesday night in the U.K. "We do not know which." Over the past few years, Hawking, who once warned that AI could spell the end of mankind, has been joined by Tesla Motors CEO Elon Musk and Microsoft founder Bill Gates in sounding the alarm on the dangers associated with it. "I am very glad someone was listening to me," Hawking, a former professor at the university, said, in reference to his and others' warnings. But Hawking's predictions for the future of AI -- and humans -- were far more optimistic than his previous doomsday comments.

This article is part of an MIT SMR initiative exploring how technology is reshaping the practice of management. The paradox of automation: Earlier this year, Facebook exorcised those pesky human editors who were introducing political bias into its Trending news list and left the job to algorithms. Now, reports Caitlin Dewey in The Washington Post, the Trending news isn't biased, but some of it is fake. Turns out the algorithms can't tell a real news story from a hoax. Facebook says it can improve its algorithms, but errors of judgment aren't the only pitfall in transferring human tasks to machines.