This paper proposes the use of Constraint Logic Programming (CLP) to model SQL queries in a data-independent abstract layer by focusing on some semantic properties for signalling possible errors in such queries. First, we define a translation from SQL to Datalog, and from Datalog to CLP, so that solving this CLP program will give information about inconsistency, tautology, and possible simplifications. We use different constraint domains which are mapped to SQL types, and propose them to cooperate for improving accuracy. Our approach leverages a deductive system that includes SQL and Datalog, and we present an implementation in this system which is currently being tested in classroom, showing its advantages and differences with respect to other approaches, as well as some performance data. This paper is under consideration for acceptance in TPLP .

Domain generalization (DG) aims to incorporate knowledge from multiple source domains into a single model that could generalize well on unseen target domains. This problem is ubiquitous in practice since the distributions of the target data may rarely be identical to those of the source data. In this paper, we propose Multidomain Discriminant Analysis (MDA) to address DG of classification tasks in general situations. MDA learns a domain-invariant feature transformation that aims to achieve appealing properties, including a minimal divergence among domains within each class, a maximal separability among classes, and overall maximal compactness of all classes. Furthermore, we provide the bounds on excess risk and generalization error by learning theory analysis. Comprehensive experiments on synthetic and real benchmark datasets demonstrate the effectiveness of MDA.

On the path to establishing a global cybersecurity framework where each enterprise shares information about malicious behavior, an important question arises. How can a machine learning representation characterizing a cyber attack on one network be used to detect similar attacks on other enterprise networks if each networks has wildly different distributions of benign and malicious traffic? We address this issue by comparing the results of naively transferring a model across network domains and using CORrelation ALignment, to our novel adversarial Siamese neural network. Our proposed model learns attack representations that are more invariant to each network's particularities via an adversarial approach. It uses a simple ranking loss that prioritizes the labeling of the most egregious malicious events correctly over average accuracy. This is appropriate for driving an alert triage workflow wherein an analyst only has time to inspect the top few events ranked highest by the model. In terms of accuracy, the other approaches fail completely to detect any malicious events when models were trained on one dataset are evaluated on another for the first 100 events. While, the method presented here retrieves sizable proportions of malicious events, at the expense of some training instabilities due in adversarial modeling. We evaluate these approaches using 2 publicly available networking datasets, and suggest areas for future research.

In spite of the rapidly increasing number of applications of machine learning in various domains, a principled and systematic approach to the incorporation of domain knowledge in the engineering process is still lacking and ad hoc solutions that are difficult to validate are still the norm in practice, which is of growing concern not only in mission-critical applications. In this note, we introduce Probabilistic Approximate Logic (PALO) as a logic based on the notion of mean approximate probability to overcome conceptual and computational difficulties inherent to strictly probabilistic logics. The logic is approximate in several dimensions. Logical independence assumptions are used to obtain approximate probabilities, but by averaging over many instances of formulas a useful estimate of mean probability with known confidence can usually be obtained. To enable efficient computational inference, the logic has a continuous semantics that reflects only a subset of the structural properties of classical logic, but this imprecision can be partly compensated by richer theories obtained by classical inference or other means. Computational inference, which refers to the construction of models and validation of logical properties, is based on Stochastic Gradient Descent (SGD) and Markov Chain Monte Carlo (MCMC) techniques and hence another dimension where approximations are involved. We also present the Logical Imagination Engine (LIME), a prototypical implementation of PALO based on TensorFlow. Albeit not limited to the biological domain, we illustrate its operation in a quite substantial bioinformatics machine learning application concerned with network synthesis and analysis in a recent DARPA project.

Nothing gets our hackles raised more than another hack that threatens vital assets. Protecting data and information along with physical assets has become the all-encompassing concern of business, government and citizens alike. Nearly every day, we see cyber criminals breach banks, credit bureaus, voting institutions, government services, medical data, and transportation systems, affecting many millions of individuals. Chances are you have personally experienced a breach. In recent years, the global cost of these attacks is estimated to be as much as $600B in funds stolen and costs to clean up the damage.

Modern data science, in its most basic form, is about understanding. While the term and process have been around for decades, it primarily existed as a subset of computer science. Today, it has grown into an independent discipline where those interested can study and major in it specifically. Modern data science employs a variety of tools and applications -- some automated -- to extract insights from digital content. It relies heavily on concepts such as mathematics, statistics, pattern recognition, predictive and probability models, machine learning and algorithmic or structured development.

Many businesses today are trying to augment and improve their customer, partner, and employee experiences by leveraging artificial intelligence and bots, yet grapple with the issue of cybersecurity. We've all heard of the numerous accounts of cybercriminals taking advantage of chat APIs, social network application vulnerabilities, and increasingly sophisticated phishing campaigns. However, the majority of cybersecurity hacks are still accomplished in a rather old-fashioned manner -- through the use of stolen credentials. As I mentioned in a recent Wall Street Journal article on the congressional hearings on AI regulation, regulators and businesses alike must take a balanced approach to AI oversight that avoids impeding innovation. Emerging technologies -- artificial intelligence (AI), internet of things (IoT), bots, and more -- demand a new level of cybersecurity that cannot be achieved with yesterday's approaches.

This afternoon, SpaceX is slated to launch its latest cargo mission from Florida for NASA, sending about 5,000 pounds of supplies to the crew on the International Space Station. For this mission, the company is employing a Dragon cargo capsule that's already been to space twice before. If successful, it'll be the first time the same Dragon has gone on a third trip to space. Packed inside the Dragon's main storage compartment are some interesting goodies and science experiments for the crew to work with over the next few months. These include a printer designed to create 3D organ-like tissues in space, as well as an experiment to culture cells taken from patients with multiple sclerosis and Parkinson's disease. The launch will also be carrying a key piece of hardware for the station itself.

Today people all around the world will be celebrating the 50th anniversary of one of humanity's greatest technological achievements: landing on the Moon. Technology has undergone immense change since 1969. The computer systems and software that took Neil Armstrong, Buzz Aldrin, and Michael Collins to our nearest celestial neighbor pale in comparison to the smartphones we carry around in our pockets today. Fifty years on, as we set our sights on a return to the Moon, as well as future human spaceflight to Mars and beyond, what are the innovations that will get us there? Research institutions and national labs across the globe are pouring hundreds of thousands of research hours into every conceivable aspect of space science.

Flight Simulator was once one of the jewels in Microsoft's crown, as close to synonymous with PC gaming as it's possible to get. The series debuted a staggering 37 years ago, pre-dating even Windows as an operating system, and demanded exacting attention from players as they guided increasingly detailed planes safely through the skies. Over the course of a dozen iterations spanning nearly four decades, the flying experience evolved from blocky cockpit views to full aerial tours with a hangar's worth of realistically modelled aircraft to get to grips with. It's been running so long that even Microsoft does not know its sales figures, but Flight Simulator has certainly been played by millions. Yet as PC gaming blossomed, becoming home to everything from competitive shooters to arthouse narrative games, Flight Simulator's star began to wane. The last major release was 2006's Microsoft Flight Simulator X (eventually revamped and repackaged for Steam in 2014), while 2012's simplified spin-off, Microsoft Flight, had an aborted take off, cancelled a mere five months after launch.