Robots aren't playing professional soccer just yet, but they can certainly help predict it! With the FIFA World Cup kicking off, San Francisco-based tech firm Unanimous A.I. has used its considerable artificial intelligence expertise to predict the outcome of the 32-team men's soccer tournament. Given that the startup has previously predicted the Super Bowl results successfully right down to the exact final score, we totally think this is worth taking seriously. "These predictions were generated using swarm A.I. technology," Louis Rosenberg, founder and CEO of Unanimous A.I., told Digital Trends. "This means it uses a unique combination of human insights and artificial intelligence algorithms, resulting in a system that is smarter than the humans or the machines could be on their own. It works by connecting a group of people over the internet using A.I. algorithms, enabling them to think together as a system, and converge upon predictions that are the optimized combination of their individual knowledge, wisdom, instincts, and intuitions."

The Brothers Grimm have been dead more than 150 years, but they recently released a new story with a little help from artificial intelligence. The Princess and the Fox was created after a group of writers, artists and developers used a programme inspired by predictive text on phones to scan the collected stories of the Brothers Grimm to suggest words and similar phrases. Human writers then took over, to help shape the AI's algorithmic suggestions into the latest Grimm fairytale. The new tale tells the story of a talking fox who helps a lowly miller's son rescue a beautiful princess from the fate of having to marry a horrible prince she does not love. But here's the thing, the Brothers Grimm didn't actually write their fairytales in the first place.

The brain of a fruit fly consists of some 100,000 different cells, and although that makes it much smaller than the human brain, it contains hundreds of different types of neurons and other cells forming a complex network, much like the human brain. To truly understand the workings of the brain, even for organisms as small as the fruit fly, we need to zoom in on each and every individual cell, explains prof. Stein Aerts (VIB-KU Leuven): "All organs and tissues are composed of many different cells that communicate with each other to perform their specific functions. Although they share the same DNA, they all express a distinct set of genes, and to understand what is really going on, we need to know which cells are doing what and when." Working with fruit flies as model organisms, the scientists took the challenge head on, immediately starting with the most complex organ of all -- the brain.

The European Commission recently announced plans to increase that funding, to make more data available for use in AI, and to work with EU member states on a strategy for deploying AI in the European economy. But at the same time, the EU's new General Data Protection Regulation (GDPR) puts tight restrictions on uses of AI that involve personal data, and EU policymakers continue to search for additional restrictions on AI to address their remaining fears. Unlike tobacco, AI has many beneficial uses, and the potential risks depend on how it is developed and used over the long-term. The irony is that if Europe over-regulates AI now, it will miss its chance for global influence over the technology's future. The commission does not see a contradiction because it believes that stringent regulation will engender consumer trust in AI. But that reasoning is flawed.

Your opportunity to shape Europe's #artificialintelligence strategy. Keep calm and "Reboot for AI revolution"Virginia Dignum on Linkedin•7mo

We propose a very simple modification of gradient descent and stochastic gradient descent. We show that when applied to a variety of machine learning models including softmax regression, convolutional neural nets, generative adversarial nets, and deep reinforcement learning, this very simple surrogate can dramatically reduce the variance and improve the accuracy of the generalization. The new algorithm, (which depends on one nonnegative parameter) when applied to non-convex minimization, tends to avoid sharp local minima. Instead it seeks somewhat flatter local (and often global) minima. The method only involves preconditioning the gradient by the inverse of a tri-diagonal matrix that is positive definite. The motivation comes from the theory of Hamilton-Jacobi partial differential equations. This theory demonstrates that the new algorithm is almost the same as doing gradient descent on a new function which (a) has the same global minima as the original function and (b) is "more convex". Again, the programming effort in doing this is minimal, in cost, complexity and effort. We implement our algorithm into both PyTorch and Tensorflow platforms, which will be made publicly available.

We consider the problem of a neural network being requested to classify images (or other inputs) without making implicit use of a "protected concept", that is a concept that should not play any role in the decision of the network. Typically these concepts include information such as gender or race, or other contextual information such as image backgrounds that might be implicitly reflected in unknown correlations with other variables, making it insufficient to simply remove them from the input features. In other words, making accurate predictions is not good enough if those predictions rely on information that should not be used: predictive performance is not the only important metric for learning systems. We apply a method developed in the context of domain adaptation to address this problem of "being right for the right reason", where we request a classifier to make a decision in a way that is entirely 'agnostic' to a given protected concept (e.g. gender, race, background etc.), even if this could be implicitly reflected in other attributes via unknown correlations. After defining the concept of an 'agnostic model', we demonstrate how the Domain-Adversarial Neural Network can remove unwanted information from a model using a gradient reversal layer.

We propose a method for simultaneously detecting shared and unshared communities in heterogeneous multilayer weighted and undirected networks. The multilayer network is assumed to follow a generative probabilistic model that takes into account the similarities and dissimilarities between the communities. We make use of a variational Bayes approach for jointly inferring the shared and unshared hidden communities from multilayer network observations. We show the robustness of our approach compared to state-of-the art algorithms in detecting disparate (shared and private) communities on synthetic data as well as on real genome-wide fibroblast proliferation dataset.

Binary classification is one of the most common problem in machine learning. It consists in predicting whether a given element is of a particular class. In this paper, a new algorithm for binary classification is proposed using a hypergraph representation. Each element to be classified is partitioned according to its interactions with the training set. For each class, the total support is calculated as a convex combination of the {\it evidence} strength of the element of the partition. The evidence measure is pre-computed using the hypergraph induced by the training set and iteratively adjusted through a training phase. It does not require structured information, each case being represented by a set of {\it agnostic information} atoms. Empirical validation demonstrates its high potential on a wide range of well-known datasets and the results are compared to the state-of-art. The time complexity is given and empirically validated. Its capacity to provide good performances without hyperparameter tuning compared to standard classification methods is studied. Finally, the limitation of the model space is discussed and some potential solutions proposed.

We consider peer review in a conference setting where there is typically an overlap between the set of reviewers and the set of authors. This overlap can incentivize strategic reviews to influence the final ranking of one's own papers. In this work, we address this problem through the lens of social choice, and present a theoretical framework for strategyproof and efficient peer review. We first present and analyze an algorithm for reviewer-assignment and aggregation that guarantees strategyproofness and a natural efficiency property called unanimity, when the authorship graph satisfies a simple property. Our algorithm is based on the so-called partitioning method, and can be thought as a generalization of this method to conference peer review settings. We then empirically show that the requisite property on the authorship graph is indeed satisfied in the ICLR-17 submission data, and further demonstrate a simple trick to make the partitioning method more practically appealing for conference peer review. Finally, we complement our positive results with negative theoretical results where we prove that under various ways of strengthening the requirements, it is impossible for any algorithm to be strategyproof and efficient.