The neural ordinary differential equation (neural ODE) model has attracted increasing attention in time series analysis for its capability to process irregular time steps, i.e., data are not observed over equally-spaced time intervals. In multi-dimensional time series analysis, a task is to conduct evolutionary subspace clustering, aiming at clustering temporal data according to their evolving low-dimensional subspace structures. Many existing methods can only process time series with regular time steps while time series are unevenly sampled in many situations such as missing data. In this paper, we propose a neural ODE model for evolutionary subspace clustering to overcome this limitation and a new objective function with subspace self-expressiveness constraint is introduced. We demonstrate that this method can not only interpolate data at any time step for the evolutionary subspace clustering task, but also achieve higher accuracy than other state-of-the-art evolutionary subspace clustering methods. Both synthetic and real-world data are used to illustrate the efficacy of our proposed method.

Or in developed nations that are less wealthy than their closest neighbors, like my native Portugal. Because of the country's modest economic size, compared to most of Western Europe, many online companies have limited (or no) presence in Portuguese. British Airways, for instance, only offers customer service in Portuguese on weekdays during business hours--and they're a global airline with enormous operations in Europe. What's more, there are almost 230 million native speakers of Portuguese worldwide, the vast majority of them in Brazil (where, yes, British Airways also offers flights). It's the sixth most spoken language in the world.

Fox News correspondent Lucas Tomlinson has the details from the Pentagon on'Special Report' Russian President Vladimir Putin inspected the country's newly unveiled "Checkmate" warplane on Tuesday. The prototype of the Sukhoi fifth-generation stealth fighter was revealed at the MAKS-2021 International Aviation and Space Salon, Reuters reported. The show opened Tuesday in Zhukovsky, outside Moscow. Fifth-generation refers to the jet's stealth characteristics, a capability to cruise at supersonic speed as well as artificial intelligence to assist the pilots, among other advanced features. "What we saw in Zhukovsky today demonstrates that the Russian aviation has a big potential for development and our aircraft making industries continue to create new competitive aircraft designs," Putin said in a speech at the show.

Chilean food-tech start-up NotCo uses artificial intelligence (AI) to identify the optimum combinations of plant proteins when creating vegan alternatives to animal-based food products. The company, set up in 2015, has attracted investment from Amazon founder Jeff Bezos and Future Positive, a US investment fund founded by Biz Stone, the co-founder of Twitter. NotCo's machine learning algorithm compares the molecular structure of dairy or meat products to plant sources, searching for proteins with similar molecular components. NotCo has a database containing over 400,000 different plants, including macronutrient breakdown and chemical composition. These factors are used to predict novel food combinations with the target flavour, texture, and functionality.

The final push for the hat trick came down to the wire. Five minutes before the deadline, the team submitted work in its third and hardest data science competition of the year in recommendation systems. Called RecSys, it's a relatively new branch of computer science that's spawned one of the most widely used applications in machine learning, one that helps millions find what they want to watch, buy and play. The team's combination of six AI models packed into the contest's limit of 20 gigabytes all of the smarts it culled from studying 750 million data points. An unusual rule in the competition said the models had to run in less than 24 hours on a single core in a cloud CPU.

We consider the task of feature selection for reconstruction which consists in choosing a small subset of features from which whole data instances can be reconstructed. This is of particular importance in several contexts involving for example costly physical measurements, sensor placement or information compression. To break the intrinsic combinatorial nature of this problem, we formulate the task as optimizing a binary mask distribution enabling an accurate reconstruction. We then face two main challenges. One concerns differentiability issues due to the binary distribution. The second one corresponds to the elimination of redundant information by selecting variables in a correlated fashion which requires modeling the covariance of the binary distribution. We address both issues by introducing a relaxation of the problem via a novel reparameterization of the logitNormal distribution. We demonstrate that the proposed method provides an effective exploration scheme and leads to efficient feature selection for reconstruction through evaluation on several high dimensional image benchmarks. We show that the method leverages the intrinsic geometry of the data, facilitating reconstruction.

We describe how answer-set programs can be used to declaratively specify counterfactual interventions on entities under classification, and reason about them. In particular, they can be used to define and compute responsibility scores as attribution-based explanations for outcomes from classification models. The approach allows for the inclusion of domain knowledge and supports query answering. A detailed example with a naive-Bayes classifier is presented.

The current preprint is an update to SituationCO v1.1 (Situation Core Ontology), which represents its new version 1.2. It specifies and defines all the terms, properties, relationships and axioms of SituationCO v1.2, being an ontology for particular and generic Situations placed at the core level in the context of a four-layered ontological architecture called FCD-OntoArch (Foundational, Core, and Domain Ontological Architecture for Sciences). This is a four-layered ontological architecture, which considers Foundational, Core, Domain and Instance levels. In turn, the domain level is split down in two sub-levels, namely: Top-domain and Low-domain ontological levels. So in fact, we can consider it to be a five-tier architecture. Ontologies at the same level can be related to each other, except for the foundational level where only ThingFO (Thing Foundational Ontology) is found. In addition, ontologies' terms and relationships at lower levels can be semantically enriched by ontologies' terms and relationships from the higher levels. Note that both ThingFO and ontologies at the core level such as SituationCO, ProcessCO, among others, are domain independent. SituationCO's terms and relationships are specialized primarily from ThingFO. It also completely reuses terms primarily from ProcessCO, ProjectCO and GoalCO ontologies. Stereotypes are the used mechanism for enriching SituationCO terms. Note that in the end of this document, we address the SituationCO vs. ThingFO non-taxonomic relationship verification matrix.

Saturation-style automated theorem provers (ATPs) based on the given clause procedure are today the strongest general reasoners for classical first-order logic. The clause selection heuristics in such systems are, however, often evaluating clauses in isolation, ignoring other clauses. This has changed recently by equipping the E/ENIGMA system with a graph neural network (GNN) that chooses the next given clause based on its evaluation in the context of previously selected clauses. In this work, we describe several algorithms and experiments with ENIGMA, advancing the idea of contextual evaluation based on learning important components of the graph of clauses.

The Internet of Things (IoT) – and what it will enable – has been a discussion point for well over a decade, but the speed, low latency and reliability of 5G promise to bring the concept to life. Network slicing will allow a wide range of product types, with distinct reliability and throughput requirements, to be run out of the same architecture, and edge computing will allow nodes to communicate directly with one another, bypassing the network's core and enhancing speed and reliability. These characteristics underpin some the most interesting projects currently making use of 5G, and have made a plethora of 5G use cases possible. Here are 18 of the best. Robots are already widely used in factories, particularly in the automotive industry.