Management actions promoted economic prosperity (positive material contributions) and agricultural abandonment (negative material contributions). The park provides nonmaterial and regulating contributions (e.g., opportunities for nature experiences, habitat creation and maintenance).

The European Commission's Joint Research Center is working on a tool that could use tweets and artificial intelligence to collect real-time data on floods. In a paper released on Arvix.org, EU scientists explain how their Social Media for Flood Risk (SMFR) prototype could help emergency responders better understand what's happening on the ground in flooded areas and determine what trouble spots might need immediate attention. The tool works in collaboration with Europe's Flood Awareness System (EFAS). When EFAS identifies areas with heightened flood risks, it triggers SMFR to begin collecting flood-related tweets from users in those areas.

A little over a year ago, we reported on the status of the Robonaut 2 on the International Space Station. Things had not gone all that well for R2 ever since an attempt had been made to install a pair of legs back in 2014, leading to an intermittent power problem that was very hard to diagnose. NASA brought Robonaut back to Earth last year for repairs, and a few weeks ago, we stopped by NASA's Johnson Space Center (JSC) in Houston, Texas, to visit the Robonaut lab and get an update on what's been happening with R2. The Robonaut lab is in Building 9 at JSC, attached to the space vehicle mockup facility. JSC's Valkyrie lives in this massive high bay, and was busy practicing a bomb disposal task (!) when we peeked in.

As agencies continue to adopt artificial intelligence and machine learning, they're working to keep their employees up to speed in new technologies and cybersecurity defenses. Lynne Parker, assistant director for AI at the White House Office of Science and Technology Policy (OSTP), acknowledged at an April 18 event hosted by the National Academy for Public Administration that AI presents "a lot of workforce challenges." Parker said that while government doesn't "at the moment" have a holistic approach toward training and preparing the workforce for technological advancements, "agencies are stepping up on their own to create opportunities for people to enroll" in IT, cyber and AI training. "Agencies are taking an active role in trying to help people learn those kinds of opportunities," she said. Because agencies understand "the kinds of skills and the kinds of work that they do, they're best suited to create these kinds of training programs," she added, citing the General Services Administration as an example of an agency that's promoting new skills.

In the era of Trumpian politics the public forum is no longer the evening news or the morning paper: it's Twitter. We decided to see which politicians were embracing the platform to discuss what we feel will be one of the most important issues of their potential presidency: artificial intelligence. What we found wasn't good. We gathered a list of every AI-related tweet 2020 Democratic hopefuls sent over the past 12 months. Our expectation was that there'd be a few thousand or so, and we'd spend the better part of the week parsing the information for insights.

European Robotics Forum, the most influential meeting of the robotics and AI community, held its 10th anniversary edition in Romania. The event was organized Under the High Patronage of the President of Romania and Under the Patronage of the Romanian Presidency of the Council of the European Union. The most advanced prototypes, high end technology projects, financed under Horizon 2020, were exhibited to be admired and analysed at JW Marriott between 20 and 22 March. Among the robots that were displayed one could find: the famous REEM-C – the humanoid robot that speaks 9 languages and that costs 1 million euro, QT – the robot especially created to help children who suffer from autism, Trimbot – the gardening robot that will help cut the roses and the bushes, as well as other prototypes that take innovation to the next level. Among the exhibitors we could also find Romanian companies developing advanced software solutions for international robotics companies.

Generative Policy-based Models aim to enable a coalition of systems, be they devices or services to adapt according to contextual changes such as environmental factors, user preferences and different tasks whilst adhering to various constraints and regulations as directed by a managing party or the collective vision of the coalition. Recent developments have proposed new architectures to realize the potential of GPMs but as the complexity of systems and their associated requirements increases, there is an emerging requirement to have scenarios and associated datasets to realistically evaluate GPMs with respect to the properties of the operating environment, be it the future battlespace or an autonomous organization. In order to address this requirement, in this paper, we present a method of applying an agile knowledge representation framework to model requirements, both individualistic and collective that enables synthetic generation of ground truth data such that advanced GPMs can be evaluated robustly in complex environments. We also release conceptual models, annotated datasets, as well as means to extend the data generation approach so that similar datasets can be developed for varying complexities and different situations.

Time Series Classification (TSC) problems are encountered in many real life data mining tasks ranging from medicine and security to human activity recognition and food safety. With the recent success of deep neural networks in various domains such as computer vision and natural language processing, researchers started adopting these techniques for solving time series data mining problems. However, to the best of our knowledge, no previous work has considered the vulnerability of deep learning models to adversarial time series examples, which could potentially make them unreliable in situations where the decision taken by the classifier is crucial such as in medicine and security. For computer vision problems, such attacks have been shown to be very easy to perform by altering the image and adding an imperceptible amount of noise to trick the network into wrongly classifying the input image. Following this line of work, we propose to leverage existing adversarial attack mechanisms to add a special noise to the input time series in order to decrease the network's confidence when classifying instances at test time. Our results reveal that current state-of-the-art deep learning time series classifiers are vulnerable to adversarial attacks which can have major consequences in multiple domains such as food safety and quality assurance.

Stratified models are models that depend in an arbitrary way on a set of selected categorical features, and depend linearly on the other features. In a basic and traditional formulation a separate model is fit for each value of the categorical feature, using only the data that has the specific categorical value. To this formulation we add Laplacian regularization, which encourages the model parameters for neighboring categorical values to be similar. Laplacian regularization allows us to specify one or more weighted graphs on the stratification feature values. For example, stratifying over the days of the week, we can specify that the Sunday model parameter should be close to the Saturday and Monday model parameters. The regularization improves the performance of the model over the traditional stratified model, since the model for each value of the categorical `borrows strength' from its neighbors. In particular, it produces a model even for categorical values that did not appear in the training data set. We propose an efficient distributed method for fitting stratified models, based on the alternating direction method of multipliers (ADMM). When the fitting loss functions are convex, the stratified model fitting problem is convex, and our method computes the global minimizer of the loss plus regularization; in other cases it computes a local minimizer. The method is very efficient, and naturally scales to large data sets or numbers of stratified feature values. We illustrate our method with a variety of examples.

We discuss a variant of `blind' community detection, in which we aim to partition an unobserved network from the observation of a (dynamical) graph signal defined on the network. We consider a scenario where our observed graph signals are obtained by filtering white noise input, and the underlying network is different for every observation. In this fashion, the filtered graph signals can be interpreted as defined on a time-varying network. We model each of the underlying network realizations as generated by an independent draw from a latent stochastic blockmodel (SBM). To infer the partition of the latent SBM, we propose a simple spectral algorithm for which we provide a theoretical analysis and establish consistency guarantees for the recovery. We illustrate our results using numerical experiments on synthetic and real data, highlighting the efficacy of our approach.