Graph kernels are kernel methods measuring graph similarity and serve as a standard tool for graph classification. However, the use of kernel methods for node classification, which is a related problem to graph representation learning, is still ill-posed and the state-of-the-art methods are heavily based on heuristics. Here, we present a novel theoretical kernel-based framework for node classification that can bridge the gap between these two representation learning problems on graphs. Our approach is motivated by graph kernel methodology but extended to learn the node representations capturing the structural information in a graph. We theoretically show that our formulation is as powerful as any positive semidefinite kernels. To efficiently learn the kernel, we propose a novel mechanism for node feature aggregation and a data-driven similarity metric employed during the training phase. More importantly, our framework is flexible and complementary to other graph-based deep learning models, e.g., Graph Convolutional Networks (GCNs). We empirically evaluate our approach on a number of standard node classification benchmarks, and demonstrate that our model sets the new state of the art.

The rise of artificial intelligence (AI) is one of the defining business opportunities for leaders today. Closely associated with it: the challenge of creating an organization that can rise to that opportunity and exploit the potential of AI at scale. Meeting this challenge requires organizations to prepare their leaders, business staff, analytics teams, and end users to work and think in new ways--not only by helping these cohorts understand how to tap into AI effectively, but also by teaching them to embrace data exploration, agile development, and interdisciplinary teamwork. Often, companies use an ad hoc approach to their talent-building efforts. They hire new workers equipped with these skills in spurts and rely on online-learning platforms, universities, and executive-level programs to train existing employees.

In this age of automation, organizations need to adapt to stay competitive. A report from Willis Tower Watson states that employers expect 17 percent of work to be automated by 2020. As technology evolves, your workforce must constantly update its skills and understand how new technology affects the nature of their jobs. But to set up employees to successfully embrace this new age of work, organizations need to future-proof their approach to learning. Rather than defining competencies designed to help their employees be successful today, they should empower their workers to focus on skills that will be critical for success tomorrow.

When you start learning a new skill, the first thing is to look at the big picture and where your would-be skills fit in the field. It gives you a context of what role you can play or are expected to play. And when the skill and field are evolving & overwhelmingly large, you are so engrossed in the details that most probably you tend to miss the purpose. In my view, to understand the big picture, ask yourself'why' more often and start with the end in your mind. The following analogy is not particularly about artificial intelligence but in general.

While recent advances in language processing with Deep Neural Networks (DNNs) present high-quality translation and classification of the texts, the Holy Grail of the language learning remains missed. That is, while humans appear capable to acquire languages in unsupervised way based on everyday conversations easily, the DNNs require extensive supervised training. Moreover, the humans are capable to acquire explainable and reasonable rules of connecting words into sentences based on grammatical rules and conversational patterns and have the grammatical and semantic categories of words well understood, with all that synonyms and homonyms. On the opposite, the very advanced DNN models remain black boxes not being understandable and inspectable. That is why we are looking for Understandable Language Processing (ULP) which would let acquisition of the language, comprehension of textual communications and production of textual messages in reasonable and transparent way.

What if there was no such thing as "real"? What if food could be made from thin air? What if electronics could last forever? These are some of the questions being tackled by Europe's top tech innovators identified by our team here at Sifted, in association with the co-working space Second Home and their Breakthrough event this month. This is not your ordinary innovator list. You may not have heard of these startups.

Jiawei Han, a professor of computer science at the University of Illinois at Urbana-Champaign, was recently named a Michael Aiken Chair, one of the University's highest awards. The endowed chair is the latest honor in Han's distinguished and pioneering career, with notable accomplishments including creating core data mining algorithms and co-authoring the textbook that is considered by many to have defined the field. Professor Han is also a busy and successful teacher with a love for "train[ing] the younger generation, whether at UIUC or all over the world on Coursera." Professor Han had three PhD students graduate in May, with one becoming a professor at Georgia Tech, one joining Google, and one joining Facebook. Students taking his classes as part of the Online Master of Computer Science in Data Science degree have an opportunity to learn from him through videos and can ask him questions directly during live office hours.

Controlled interacting particle systems such as the ensemble Kalman filter (EnKF) and the feedback particle filter (FPF) are numerical algorithms to approximate the solution of the nonlinear filtering problem in continuous time. The distinguishing feature of these algorithms is that the Bayesian update step is implemented using a feedback control law. It has been noted in the literature that the control law is not unique. This is the main problem addressed in this paper. To obtain a unique control law, the filtering problem is formulated here as an optimal transportation problem. An explicit formula for the (mean-field type) optimal control law is derived in the linear Gaussian setting. Comparisons are made with the control laws for different types of EnKF algorithms described in the literature. Via empirical approximation of the mean-field control law, a finite-$N$ controlled interacting particle algorithm is obtained. For this algorithm, the equations for empirical mean and covariance are derived and shown to be identical to the Kalman filter. This allows strong conclusions on convergence and error properties based on the classical filter stability theory for the Kalman filter. It is shown that, under certain technical conditions, the mean squared error (m.s.e.) converges to zero even with a finite number of particles. A detailed propagation of chaos analysis is carried out for the finite-$N$ algorithm. The analysis is used to prove weak convergence of the empirical distribution as $N\rightarrow\infty$. For a certain simplified filtering problem, analytical comparison of the m.s.e. with the importance sampling-based algorithms is described. The analysis helps explain the favorable scaling properties of the control-based algorithms reported in several numerical studies in recent literature.

Researchers at CHILI Lab (Ecole Polytechnique Fédérale de Lausanne) in Switzerland and GAIPS Lab (University of Lisbon) in Portugal have recently developed an autonomous system designed to assist children in improving their handwriting skills. The system they created, presented in a paper published in Springer's International Journal of Social Robotics, entails the use of a social robot in one-to-one learning sessions with children. For some children, handwriting can be a difficult skill to acquire, yet it is a fundamental stepping stone in their academic path. In fact, poor handwriting can negatively affect a child's academic performance, self-esteem and learning motivation. To master handwriting, a child needs to learn to coordinate cognitive, motor and perceptual abilities, thus he/she might also require a considerable amount of practice.

Artificial intelligence is having a huge impact on education, transforming the sector in many positive ways and its impact is growing. In fact, the artificial intelligence sector in the U.S. education market is expected to grow 47.5% between 2017 and 2021 according to the latest market research report by Technavio. Artificial intelligence is changing how teachers are doing their jobs and how students are learning and studying. AI makes personalized learning possible, can assist teachers with curriculum adaptationand streamline administrative tasks. Now, scientists are trying to find out if the technology can be leveraged to measure student engagement.