The Volkswagen Group continues to drive the digital transformation forward and is expanding its competence with regard to artificial intelligence (AI), not only with a view to using this key technology in the field of autonomous driving and in production, but also to speed up corporate processes. To that end, Prof. Patrick van der Smagt will be joining the team of AI experts and data scientists at the Volkswagen Group's Data Lab in Munich from October, and will lead the AI team. Van der Smagt is Associate Professor at Technische Universität München and one of Europe's leading experts in deep learning and robotics. Dr. Martin Hofmann, Chief Information Officer at the Volkswagen Group, commented: "We are delighted that Professor van der Smagt will be joining the Data Lab. This shows how attractive the Volkswagen Group's IT labs are for distinguished scientists and experts in artificial intelligence. Top specialists from Volkswagen, industry partners and the academic world as well as startups jointly develop and test forward-looking ideas in our labs – working in an environment similar to Silicon Valley."

Most fintech companies offer better interfaces for handling money (bank accounts, loans, payments, etc.), but quite a few startups are also offering a solution for a much older need than how to make money. Israel company I Know First, managed by CEO Yaron Golgher, is one of these. The company's main product is an algorithm that provides a forecast for three thousand different investment instruments, including shares, commodities, interest rates, foreign currency, exchange traded funds (ETFs), and global indices. "The algorithm rates all the investment instruments, and singles out investment opportunities in the capital market on a daily basis, according to the pricing anomaly it finds," Golgher says in a "Globes" interview. Golgher: "The algorithm is self-learning. It is based on purely quantitative values, not reading news or any kind of analysis. There is no human factor here. The algorithm uses artificial intelligence, an area in which huge companies like Apple Computers, Google, and Facebook have recently been making massive investments. The algorithm was developed by our development team, headed by cofounder and CTO Dr. Lipa Roitman. Roitman is a scientist from the Weizmann Institute of Science with over 20 years of experience in the specific field of artificial intelligence, machine learning, and algorithms. The forecast is given for a period of time. What is interesting is that for every forecast, the algorithm also assigns a probability that the forecast will be fulfilled. Every customer can receive a forecast according to his investment preferences. For example, a person investing in technology shares can receive the best opportunities in this segment, a customer investing in commodities will receive the best opportunities in the commodities market, etc." "We have expanded this year to 12 new countries, including the US and Europe, with an emphasis on Italy and France, and Russia, too. We work with Latin America, especially Brazil. The business model is based on access to the algorithm to a varying extent, according to the customer's size."

Now we can use the function f f(x), which we call the target function, as the proxy for the conditional distribution.

First--try some of the introductory tutorial/competitions. Those get your feet wet. Then just jump head first into a competition. Try and be active on the forums. I have found that the best way to learn is just struggle with it (in most anything--I faked my way into a DB engineer once, 2 years later I was teaching the course on SQL at a Fortune 100 company--I had my share of run-ins with the Admin though--we were on a first name basis)).

State-space smoothing has found many applications in science and engineering. Under linear and Gaussian assumptions, smoothed estimates can be obtained using efficient recursions, for example Rauch-Tung-Striebel and Mayne-Fraser algorithms. Such schemes are equivalent to linear algebraic techniques that minimize a convex quadratic objective function with structure induced by the dynamic model. These classical formulations fall short in many important circumstances. For instance, smoothers obtained using quadratic penalties can fail when outliers are present in the data, and cannot track impulsive inputs and abrupt state changes. Motivated by these shortcomings, generalized Kalman smoothing formulations have been proposed in the last few years, replacing quadratic models with more suitable, often nonsmooth, convex functions. In contrast to classical models, these general estimators require use of iterated algorithms, and these have received increased attention from control, signal processing, machine learning, and optimization communities. In this survey we show that the optimization viewpoint provides the control and signal processing community great freedom in the development of novel modeling and inference frameworks for dynamical systems. We discuss general statistical models for dynamic systems, making full use of nonsmooth convex penalties and constraints, and providing links to important models in signal processing and machine learning. We also survey optimization techniques for these formulations, paying close attention to dynamic problem structure. Modeling concepts and algorithms are illustrated with numerical examples.

Support Vector Data Description (SVDD) is a popular outlier detection technique which constructs a flexible description of the input data. SVDD computation time is high for large training datasets which limits its use in big-data process-monitoring applications. We propose a new iterative sampling-based method for SVDD training. The method incrementally learns the training data description at each iteration by computing SVDD on an independent random sample selected with replacement from the training data set. The experimental results indicate that the proposed method is extremely fast and provides a good data description .

Monte Carlo (MC) methods are widely used for Bayesian inference and optimization in statistics, signal processing and machine learning. A well-known class of MC methods are Markov Chain Monte Carlo (MCMC) algorithms. In order to foster better exploration of the state space, specially in high-dimensional applications, several schemes employing multiple parallel MCMC chains have been recently introduced. In this work, we describe a novel parallel interacting MCMC scheme, called {\it orthogonal MCMC} (O-MCMC), where a set of "vertical" parallel MCMC chains share information using some "horizontal" MCMC techniques working on the entire population of current states. More specifically, the vertical chains are led by random-walk proposals, whereas the horizontal MCMC techniques employ independent proposals, thus allowing an efficient combination of global exploration and local approximation. The interaction is contained in these horizontal iterations. Within the analysis of different implementations of O-MCMC, novel schemes in order to reduce the overall computational cost of parallel multiple try Metropolis (MTM) chains are also presented. Furthermore, a modified version of O-MCMC for optimization is provided by considering parallel simulated annealing (SA) algorithms. Numerical results show the advantages of the proposed sampling scheme in terms of efficiency in the estimation, as well as robustness in terms of independence with respect to initial values and the choice of the parameters.

Learning based on networks of real neurons, and by extension biologically inspired models of neural networks, has yet to find general learning rules leading to widespread applications. In this paper, we argue for the existence of a principle allowing to steer the dynamics of a biologically inspired neural network. Using carefully timed external stimulation, the network can be driven towards a desired dynamical state. We term this principle "Learning by Stimulation Avoidance" (LSA). We demonstrate through simulation that the minimal sufficient conditions leading to LSA in artificial networks are also sufficient to reproduce learning results similar to those obtained in biological neurons by Shahaf and Marom [1]. We examine the mechanism's basic dynamics in a reduced network, and demonstrate how it scales up to a network of 100 neurons. We show that LSA has a higher explanatory power than existing hypotheses about the response of biological neural networks to external simulation, and can be used as a learning rule for an embodied application: learning of wall avoidance by a simulated robot. The surge in popularity of artificial neural networks is mostly directed to disembodied models of neurons with biologically irrelevant dynamics: to the authors' knowledge, this is the first work demonstrating sensory-motor learning with random spiking networks through pure Hebbian learning.

However, Oren Etzioni, professor of Computer Science at the University of Washington and CEO of the Allen Institute for Artificial Intelligence, argues that such headlines are in fact strongly influenced by the work of one man: professor Nick Bostrom of the Faculty of Philosophy at Oxford University, author of the bestselling treatise Superintelligence: Paths, Dangers, and Strategies. Essentially, Bostrom claims that if machine brains surpass human brains in general intelligence, the resultant new'superintelligence' could replace humans as the dominant lifeform on Earth. Furthermore, according to his findings, there's a 10-percent probability that human-level AI will be attained by 2022, a 50-percent probability that this feat will be achieved by 2040, and 90-percent probability that such an entity will be created by 2075. However, in his article published in the MIT Technology Review magazine Etzioni points out that Bostrom's main source of data is an aggregate of four different surveys of groups, including participants of the Philosophy and Theory of AI conference that was held in 2011 in Thessaloniki, and members of the Greek Association for Artificial Intelligence. Furthermore, it appears that Bostrom didn't provide the response rates or the phrasing of questions used during those surveys, and neither did he account for the reliance on data collected in Greece.

Science has proven what many a beer drinker probably already knows – drinking beer makes you nicer and more likely to be attracted to others. That's according to a new study conducted by researchers in Switzerland, the BBC reports, where a team from University Hospital in Basel conducted tests with 60 men and women. Researchers had an equal number of both sexes engage in a wide variety of tasks, including face recognition, empathy and sexual arousal tests while drinking either regular beer with alcohol or a non-alcoholic brew. The study found that those who drank beer with alcohol wanted to be around others and drinking also made people more talkative. The desire to be more social was also more pronounced in women and in those study participants with higher initial inhibitions.