Neural networks have been used prominently in several machine learning and statistics applications. In general, the underlying optimization of neural networks is non-convex which makes their performance analysis challenging. In this paper, we take a novel approach to this problem by asking whether one can constrain neural network weights to make its optimization landscape have good theoretical properties while at the same time, be a good approximation for the unconstrained one. For two-layer neural networks, we provide affirmative answers to these questions by introducing Porcupine Neural Networks (PNNs) whose weight vectors are constrained to lie over a finite set of lines. We show that most local optima of PNN optimizations are global while we have a characterization of regions where bad local optimizers may exist. Moreover, our theoretical and empirical results suggest that an unconstrained neural network can be approximated using a polynomially-large PNN.

Understanding player behavior is fundamental in game data science. Video games evolve as players interact with the game, so being able to foresee player experience would help to ensure a successful game development. In particular, game developers need to evaluate beforehand the impact of in-game events. Simulation optimization of these events is crucial to increase player engagement and maximize monetization. We present an experimental analysis of several methods to forecast game-related variables, with two main aims: to obtain accurate predictions of in-app purchases and playtime in an operational production environment, and to perform simulations of in-game events in order to maximize sales and playtime. Our ultimate purpose is to take a step towards the data-driven development of games. The results suggest that, even though the performance of traditional approaches such as ARIMA is still better, the outcomes of state-of-the-art techniques like deep learning are promising. Deep learning comes up as a well-suited general model that could be used to forecast a variety of time series with different dynamic behaviors.

By exploiting the property that the RBM log-likelihood function is the difference of convex functions, we formulate a stochastic variant of the difference of convex functions (DC) programming to minimize the negative log-likelihood. Interestingly, the traditional contrastive divergence algorithm is a special case of the above formulation and the hyperparameters of the two algorithms can be chosen such that the amount of computation per mini-batch is identical. We show that for a given computational budget the proposed algorithm almost always reaches a higher log-likelihood more rapidly, compared to the standard contrastive divergence algorithm. Further, we modify this algorithm to use the centered gradients and show that it is more efficient and effective compared to the standard centered gradient algorithm on benchmark datasets.

DeepMind, the Google sibling focusing on artificial intelligence, has announced the launch of an'Ethics and Society' unit to study the impact of new technologies on society. The announcement by the London-based group acquired by Google parent Alphabet is the latest effort in the tech sector to ease concerns that robotics and artificial intelligence will veer out of human control. 'As scientists developing AI technologies, we have a responsibility to conduct and support open research and investigation into the wider implications of our work,' said a blog post announcing the launch Tuesday by DeepMind's Verity Harding and Sean Legassick. DeepMind, the Google sibling focusing on artificial intelligence, has announced the launch of an'Ethics and Society' unit to study the impact of new technologies on society. Google's DeepMind AI relies on artificial neural networks, which try to simulate the way the brain works in order to learn.

Part 3. Here's a look at industry specific companies that utilise various forms of artificial intelligence to solve some really interesting and particular problems for different markets. Basket -- e-commerce shopping cart chatbot Choice.ai AltSchool -- a platform made to improve learning capabailities Content Technologies (CTI) -- research and development company Coursera -- online courses from top universities Gradescope -- streamlines the tedious parts of grading Hugh -- helps library users find any book quickly Ivy.ai -- customer service chatbot for higher education Knewton -- personalised learning for high and primary schools Volley -- makes training and development more engaging and effective AlphaSense -- highly intelligent search functionality Alta5 -- scriptable trading automation for your online brokerage account Analytic.ai Atomwise -- for novel small molecule discovery Babylon -- online doctor consultations using AI BuddiHealth -- helps improve process, payment systems and costs with RCM Behold.ai Imagia -- helps detect changes in cancer early Kuznech -- computer vision products range Lunit Inc. -- a range of medical imaging software Zebra Medical Vision -- medical imaging to help physicians and practitioners Cape Analytics -- identify property attributes at scale for underwriting Underwrite.ai

Researchers demo how deep neural networks can be trained to generate passwords better than the best password-cracking tools. Researchers at the Stevens Institute of Technology in New York, and the New York Institute of Technology have devised what they claim is a highly effective way to guess passwords using a deep learning tool called Generative Adversarial Networks (GANs). Tests of the'PassGAN' technique, as the researchers are calling it, show the method to be an improvement over state-of-the-art, rules-based password guessing tools such as HashCat and John the Ripper, the researchers said in a recently published technical paper. In their experiments the researchers were able to match nearly 47% - or some 2,774,269 out of 5,919,936 passwords - from a testing set comprised of real user passwords that were publicly leaked after a 2010 data breach at RockYou. Overall, the evaluations showed PassGAN outperforming John the Ripper by a factor of two, and being at least as competitive with passwords generated using the best rules from HashCat. When the output from PassGAN was combined with HashCat output the researchers could match about 24% more passwords than generated by HashCat alone.

Machine learning is now mainstream. And given the success companies see deriving value from the vast amount of available data, everyone wants in. But while the thought of machine learning can seem overwhelming, it's not magic, and the basic concepts are fairly simple. Here I'll give you a foundation for understanding the ideas behind some of the most popular machine learning algorithms. Devised by Leo Breiman in 2001, a random forest is a simple, yet powerful algorithm comprised of an ensemble (or collection)of independently trained decision trees.

Google's DeepMind artificial intelligence (AI) division has established a new research group to learn more about the ethical questions posed by the … Read more: DeepMind wants to answer the big ethical questions posed by AI

As co-founder and CEO of MarianaIQ, Venkat Nagaswamy brings a long and diverse background in high technology to bear on applying artificial intelligence and Deep Learning to help marketers make account-based marketing (ABM) an at-scale reality. "Big Kat," as he was nicknamed by friends and colleagues, has led teams in creating analytics, technology and business development solutions at McKinsey, Juniper Networks and GE Plastics, among others. He's worked in enterprise and digital consumer hardware, SaaS, corporate and business unit strategy, market entry strategy, product development, marketing planning and more, allowing him to understand martech challenges from both the CTO and CMO's point of view. A proud graduate of the University of Michigan and the Georgia Institute of Technology, he holds an MBA and a Master's in Aerospace Engineering.

Artificial intelligence (AI) is expected to have a monumental impact on society.