Deep neural networks are a family of computational models that have led to a dramatical improvement of the state of the art in several domains such as image, voice or text analysis. These methods provide a framework to model complex, non-linear interactions in large datasets, and are naturally suited to the analysis of hierarchical data such as, for instance, longitudinal data with the use of recurrent neural networks. In the other hand, cohort studies have become a tool of importance in the research field of epidemiology. In such studies, variables are measured repeatedly over time, to allow the practitioner to study their temporal evolution as trajectories, and, as such, as longitudinal data. This paper investigates the application of the advanced modelling techniques provided by the deep learning framework in the analysis of the longitudinal data provided by cohort studies. Methods: A method for visualizing and clustering longitudinal dataset is proposed, and compared to other widely used approaches to the problem on both real and simulated datasets. Results: The proposed method is shown to be coherent with the preexisting procedures on simple tasks, and to outperform them on more complex tasks such as the partitioning of longitudinal datasets into non-spherical clusters. Conclusion: Deep artificial neural networks can be used to visualize longitudinal data in a low dimensional manifold that is much simpler to interpret than traditional longitudinal plots are. Consequently, practitioners should start considering the use of deep artificial neural networks for the analysis of their longitudinal data in studies to come.

We propose an extension of the canonical polyadic (CP) tensor model where one of the latent factors is allowed to vary through data slices in a constrained way. The components of the latent factors, which we want to retrieve from data, can vary from one slice to another up to a diffeomorphism. We suppose that the diffeomorphisms are also unknown, thus merging curve registration and tensor decomposition in one model, which we call registered CP. We present an algorithm to retrieve both the latent factors and the diffeomorphism, which is assumed to be in a parametrized form. At the end of the paper, we show simulation results comparing registered CP with other models from the literature.

Recently, a new multi-step temporal learning algorithm, called $Q(\sigma)$, unifies $n$-step Tree-Backup (when $\sigma=0$) and $n$-step Sarsa (when $\sigma=1$) by introducing a sampling parameter $\sigma$. However, similar to other multi-step temporal-difference learning algorithms, $Q(\sigma)$ needs much memory consumption and computation time. Eligibility trace is an important mechanism to transform the off-line updates into efficient on-line ones which consume less memory and computation time. In this paper, we further develop the original $Q(\sigma)$, combine it with eligibility traces and propose a new algorithm, called $Q(\sigma ,\lambda)$, in which $\lambda$ is trace-decay parameter. This idea unifies Sarsa$(\lambda)$ (when $\sigma =1$) and $Q^{\pi}(\lambda)$ (when $\sigma =0$). Furthermore, we give an upper error bound of $Q(\sigma ,\lambda)$ policy evaluation algorithm. We prove that $Q(\sigma,\lambda)$ control algorithm can converge to the optimal value function exponentially. We also empirically compare it with conventional temporal-difference learning methods. Results show that, with an intermediate value of $\sigma$, $Q(\sigma ,\lambda)$ creates a mixture of the existing algorithms that can learn the optimal value significantly faster than the extreme end ($\sigma=0$, or $1$).

The mathematical problem of optimal mass transport has a long history dating back to its introduction in Monge [10], with key contributions by Kantorovivc [6] and Kantorovivc & Rubinvsteuin [7]. It has recently received increased interest due to numerous applications in machine learning; see, e.g., the recent overview of Kolouri, Park, Thorpe, Slepcev & Rohde [9] and the references therein. In a nutshell, the (discrete) problem of optimal transport in its Kantorovich form is to compute for given mass distributions a and b with equal mass a transport plan, i.e., an assignment of how much mass of a at some point should be moved to another point to match the mass in b. This should be done in a way such that some transport cost (usually proportional to the amount of mass and dependent on the distance) is minimized. This leads to a linear optimization problem which has been well studied, but its application in machine learning has been problematic due to large memory requirement and long run time.

Graph representations offer powerful and intuitive ways to describe data in a multitude of application domains. Here, we consider stochastic processes generating graphs and propose a methodology for detecting changes in stationarity of such processes. The methodology is general and considers a process generating attributed graphs with a variable number of vertices/edges, without the need to assume one-to-one correspondence between vertices at different time steps. The methodology acts by embedding every graph of the stream into a vector domain, where a conventional multivariate change detection procedure can be easily applied. We ground the soundness of our proposal by proving several theoretical results. In addition, we provide a specific implementation of the methodology and evaluate its effectiveness on several detection problems involving attributed graphs representing biological molecules and drawings. Experimental results are contrasted with respect to suitable baseline methods, demonstrating the effectiveness of our approach.

It is important to properly take care of your body and health. This can be achieved by many ways. But one important way to take care of our health is by gaining healthy food. But nowadays in our busy life because of the tight schedules, people tend to approach fast-food instead of preparing healthy meals at home which leads to severe diseases like obesity and type 2 diabetes. Concerned about these issues and noticing the increased rate of consumers for healthy eating, there are some cooking robots which enable us to cook the food.

My guest Robert Draper says one of the greatest threats to our democracy is gerrymandering, in which the party in power in a state redraws the map of election districts to give the advantage to that party's candidates. Since districts are redrawn only every 10 years following the census, gerrymandering can almost guarantee that the majority party will stay in power. There are a couple of gerrymandering cases currently before the Supreme Court. Draper has reported on gerrymandering, and we'll talk about that a little later. First, we're going to talk about his new article "They Are Watching You - And Everything Else On The Planet" published in this month's National Geographic. It's about state-of-the-art surveillance from closed-circuit TV to drones and satellites and the questions these surveillance technologies raise about privacy. As part of his research, he spent time in surveillance control rooms in London. And he went to a tech company in San Francisco whose mission is to image the entire Earth every day. Draper is a contributing writer for National Geographic and a writer at large for The New York Times Magazine. So let's start with surveillance. Why did you choose England as the place to report on surveillance? ROBERT DRAPER: Well, England has become kind of an obvious focal point to talk about surveillance. It's become, in a way, a petri dish for the subject, I suppose, for a couple reasons. First of all, the U.K. is where George Orwell wrote his dystopian classic "1984" back in 1949 when the totalitarianism of Nazi Germany and the USSR were his prime reference points.

New research highlights some of the most important actions available to executives. McKinsey research has long demonstrated the wide gap between productivity levels in different countries. Research in 2015, for example, suggested that if the degree of productivity dispersion among the bottom 75 percent of UK firms matched that of Germany, the United Kingdom would be more than £100 billion better off annually as measured by incremental gross value added (GVA). This analysis also showed that a major reason for that discrepancy is the United Kingdom's relatively slower diffusion of digital technologies and proven business practices among the bulk of its business population. We set out recently to investigate what drives, and holds back, the diffusion of technology-enabled business practices, using a mix of academic literature, studies from multinational organizations such as the Organisation for Economic Co-operation and Development (OECD) and the World Economic Forum, and in-depth interviews with business leaders and other experts.

The risks posed by intelligent devices will soon surpass the magnitude of those associated with natural disasters. Tens of billions of connected sensors are being embedded in everything ranging from industrial robots and safety systems to self-driving cars and refrigerators. At the same time, the capabilities of artificial intelligence (AI) algorithms are evolving rapidly. Our growing reliance on so many intelligent, connected devices is opening up the possibility of global-scale shutdowns. The good news is that natural disasters themselves, which Munich Re says caused $330 billion in economic losses globally in 2017, provide a template for how to mitigate the growing and catastrophic risk posed by AI.

Researchers from the Vienna University of Technology (VUT) have put a brain on a circuit board -- specifically, the brain of the nematode C. elegans. They are now training it to perform tasks without a single line of human-written code. Image credits PROZEISS Microscopy / Flickr. But in one respect, this little nematode is unique and uniquely valuable for science -- it's the only living being whose neural system has been fully analyzed and mapped. In other words, its brain can be recreated as a circuit -- either onto a circuit board or one simulated with software -- without losing any of its function.