Twenty-four EU countries pledged to band together to form a "European approach" to artificial intelligence in a bid to compete with American and Asian tech giants. Ministers signed a declaration on Tuesday (10 April) saying they will consider putting public research funding into AI, but did not promise a specific amount of dedicated new investments. All EU member states except for Cyprus, Romania, Croatia and Greece vowed to "modernise national policies" as part of an effort to develop large-scale AI research. One Commission official said the four EU countries that did not sign were not opposed to the initiative but might still need formal approval before signing. Norway also signed the declaration.

Stephen Hawking is one of the most respected minds in science today. He often speaks on a wide range of topics both within and outside of his particular expertise in theoretical physics. Some of Hawking's most discussed topics including the search for alien life, climate change, artificial intelligence (AI), and how all of these things, and more, are going to spell the end of humanity once and for all. Speaking at an event at Cambridge University last year Hawking said, "Our earth is becoming too small for us, global population is increasing at an alarming rate and we are in danger of self-destructing." He recognizes the pessimism of his assertations.

On a Swiss roll: Sophia, a human-like robot developed by Hanson Robotics, at AI for Good, a June 2017 global summit in Geneva for government, charity, industry and technology leaders to discuss AI's ethical, technical, societal and policy issues. The Alpine nation is now "the Silicon Valley of robotics," according to Chris Anderson, chief executive of 3D Robotics. Zurich boasts Google's largest campus outside California, employing nearly 2,500 engineers, including more than 250 artificial intelligence specialists, with capacity to grow the total workforce there to 5,000. The Swiss office of Disney meanwhile carried out the computer graphics working for the movie-maker's Frozen super-hit. And the key personnel and technologies of Deepmind, the artificial intelligence firm acquired by Google for $500 million just four years after its formation, emanate from Lugano IDSIA Lab, a research institute ranked in the world's top ten for AI. Switzerland has emerged as a serious competitor to California for the technologies, people and funding that will power the world's fourth socio-economic revolution.

You can create skimmers with the formula syntax from rlang! You can now control the object name output for topojson_write, and there's now an analog of geojson_sp for sf (geojson_sf) We accept community contributed packages via our onboarding system - an open software review system, sorta like scholarly paper review, but way better. We'll highlight newly onboarded packages here. A huge thanks to our reviewers, who do a lot of work reviewing (see the blog post on our review system), and the authors of the packages! If you want to be a reviewer fill out this short form, and we'll ping you when there's a submission that fits in your area of expertise.

Facing hours of sharp questioning in Congress over his company's privacy practices and its response to Russian election meddling, Facebook founder Mark Zuckerberg repeatedly argued on Tuesday that his company is engaged in an "arms race" with Moscow's intelligence agencies. Zuckerberg said that artificial intelligence represents the best solution to misinformation on Facebook, but won't be ready for another five to ten years. In the run-up to the 2016 election, Russian intelligence operatives used Facebook and other online platforms to spread divisive content generally aimed at boosting President Donald Trump's electoral chances. On Tuesday, a remorseful Zuckerberg admitted that his company's response was slow, calling it one of his "greatest regrets." Zuckerberg argued that the highly sophisticated Russian approach to spreading its influence online has left his company at a distinct disadvantage.

In this paper, we investigate cost-aware joint learning and optimization for multi-channel opportunistic spectrum access in a cognitive radio system. We investigate a discrete time model where the time axis is partitioned into frames. Each frame consists of a sensing phase, followed by a transmission phase. During the sensing phase, the user is able to sense a subset of channels sequentially before it decides to use one of them in the following transmission phase. We assume the channel states alternate between busy and idle according to independent Bernoulli random processes from frame to frame. To capture the inherent uncertainty in channel sensing, we assume the reward of each transmission when the channel is idle is a random variable. We also associate random costs with sensing and transmission actions. Our objective is to understand how the costs and reward of the actions would affect the optimal behavior of the user in both offline and online settings, and design the corresponding opportunistic spectrum access strategies to maximize the expected cumulative net reward (i.e., reward-minus-cost). We start with an offline setting where the statistics of the channel status, costs and reward are known beforehand. We show that the the optimal policy exhibits a recursive double threshold structure, and the user needs to compare the channel statistics with those thresholds sequentially in order to decide its actions. With such insights, we then study the online setting, where the statistical information of the channels, costs and reward are unknown a priori. We judiciously balance exploration and exploitation, and show that the cumulative regret scales in O(log T). We also establish a matched lower bound, which implies that our online algorithm is order-optimal. Simulation results corroborate our theoretical analysis.

In this paper, we extend a symbolic association framework for being able to handle missing elements in multimodal sequences. The general scope of the work is the symbolic associations of object-word mappings as it happens in language development in infants. In other words, two different representations of the same abstract concepts can associate in both directions. This scenario has been long interested in Artificial Intelligence, Psychology, and Neuroscience. In this work, we extend a recent approach for multimodal sequences (visual and audio) to also cope with missing elements in one or both modalities. Our method uses two parallel Long Short-Term Memories (LSTMs) with a learning rule based on EM-algorithm. It aligns both LSTM outputs via Dynamic Time Warping (DTW). We propose to include an extra step for the combination with the max operation for exploiting the common elements between both sequences. The motivation behind is that the combination acts as a condition selector for choosing the best representation from both LSTMs. We evaluated the proposed extension in the following scenarios: missing elements in one modality (visual or audio) and missing elements in both modalities (visual and sound). The performance of our extension reaches better results than the original model and similar results to individual LSTM trained in each modality.

We propose Rademacher complexity bounds for multi-class classifiers trained with a two-step semi-supervised model. In the first step, the algorithm partitions the partially labeled data and then identifies dense clusters containing κ predominant classes using the labeled training examples such that the proportion of their non-predominant classes is below a fixed threshold stands for clustering consistency. In the second step, a classifier is trained by minimizing a margin empirical loss over the labeled training set and a penalization term measuring the disability of the learner to predict the κ predominant classes of the identified clusters. The resulting data-dependent generalization error bound involves the margin distribution of the classifier, the stability of the clustering technique used in the first step and Rademacher complexity terms corresponding to partially labeled training data. Our theoretical result exhibit convergence rates extending those proposed in the literature for the binary case, and experimental results on different multi-class classification problems show empirical evidence that supports the theory.

We propose a modification of linear discriminant analysis, referred to as compressive regularized discriminant analysis (CRDA), for analysis of high-dimensional datasets. CRDA is specially designed for feature elimination purpose and can be used as gene selection method in microarray studies. CRDA lends ideas from $\ell_{q,1}$ norm minimization algorithms in the multiple measurement vectors (MMV) model and utilizes joint-sparsity promoting hard thresholding for feature elimination. A regularization of the sample covariance matrix is also needed as we consider the challenging scenario where the number of features (variables) is comparable or exceeding the sample size of the training dataset. A simulation study and four examples of real-life microarray datasets evaluate the performances of CRDA based classifiers. Overall, the proposed method gives fewer misclassification errors than its competitors, while at the same time achieving accurate feature selection.

Deep Learning Library (DLL) is a new library for machine learning with deep neural networks that focuses on speed. It supports feed-forward neural networks such as fully-connected Artificial Neural Networks (ANNs) and Convolutional Neural Networks (CNNs). It also has very comprehensive support for Restricted Boltzmann Machines (RBMs) and Convolutional RBMs. Our main motivation for this work was to propose and evaluate novel software engineering strategies with potential to accelerate runtime for training and inference. Such strategies are mostly independent of the underlying deep learning algorithms. On three different datasets and for four different neural network models, we compared DLL to five popular deep learning frameworks. Experimentally, it is shown that the proposed framework is systematically and significantly faster on CPU and GPU. In terms of classification performance, similar accuracies as the other frameworks are reported.