Though anomaly detection (AD) can be viewed as a classification problem (nominal vs. anomalous) it is usually treated in an unsupervised manner since one typically does not have access to, or it is infeasible to utilize, a dataset that sufficiently characterizes what it means to be "anomalous." In this paper we present results demonstrating that this intuition surprisingly does not extend to deep AD on images. For a recent AD benchmark on ImageNet, classifiers trained to discern between normal samples and just a few (64) random natural images are able to outperform the current state of the art in deep AD. We find that this approach is also very effective at other common image AD benchmarks. Experimentally we discover that the multiscale structure of image data makes example anomalies exceptionally informative.

In recent years, considerable progress has been made on improving the interpretability of machine learning models. This is essential, as complex deep learning models with millions of parameters produce state of the art results, but it can be nearly impossible to explain their predictions. While various explainability techniques have achieved impressive results, nearly all of them assume each data instance to be independent and identically distributed (iid). This excludes relational models, such as Statistical Relational Learning (SRL), and the recently popular Graph Neural Networks (GNNs), resulting in few options to explain them. While there does exist one work on explaining GNNs, GNN-Explainer, they assume access to the gradients of the model to learn explanations, which is restrictive in terms of its applicability across non-differentiable relational models and practicality. In this work, we develop RelEx, a model-agnostic relational explainer to explain black-box relational models with only access to the outputs of the black-box. RelEx is able to explain any relational model, including SRL models and GNNs. We compare RelEx to the state-of-the-art relational explainer, GNN-Explainer, and relational extensions of iid explanation models and show that RelEx achieves comparable or better performance, while remaining model-agnostic.

It is a common phenomenon that for high-dimensional and nonparametric statistical models, rate-optimal estimators balance squared bias and variance. Although this balancing is widely observed, little is known whether methods exist that could avoid the trade-off between bias and variance. We propose a general strategy to obtain lower bounds on the variance of any estimator with bias smaller than a prespecified bound. This shows to which extent the bias-variance trade-off is unavoidable and allows to quantify the loss of performance for methods that do not obey it. The approach is based on a number of abstract lower bounds for the variance involving the change of expectation with respect to different probability measures as well as information measures such as the Kullback-Leibler or chi-square divergence. Some of these inequalities rely on a new concept of information matrices. In a second part of the article, the abstract lower bounds are applied to several statistical models including the Gaussian white noise model, a boundary estimation problem, the Gaussian sequence model and the high-dimensional linear regression model. For these specific statistical applications, different types of bias-variance trade-offs occur that vary considerably in their strength. For the trade-off between integrated squared bias and integrated variance in the Gaussian white noise model, we propose to combine the general strategy for lower bounds with a reduction technique. This allows us to reduce the original problem to a lower bound on the bias-variance trade-off for estimators with additional symmetry properties in a simpler statistical model. To highlight possible extensions of the proposed framework, we moreover briefly discuss the trade-off between bias and mean absolute deviation.

The twin support vector machine and its extensions have made great achievements in dealing with binary classification problems, however, which is faced with some difficulties such as model selection and solving multi-classification problems quickly. This paper is devoted to the fast regularization parameter tuning algorithm for the twin multi-class support vector machine. A new sample dataset division method is adopted and the Lagrangian multipliers are proved to be piecewise linear with respect to the regularization parameters by combining the linear equations and block matrix theory. Eight kinds of events are defined to seek for the starting event and then the solution path algorithm is designed, which greatly reduces the computational cost. In addition, only few points are combined to complete the initialization and Lagrangian multipliers are proved to be 1 as the regularization parameter tends to infinity. Simulation results based on UCI datasets show that the proposed method can achieve good classification performance with reducing the computational cost of grid search method from exponential level to the constant level.

This work is done as part of a research master's thesis project. The goal is to generate SPARQL queries based on user-supplied keywords to query RDF graphs. To do this, we first transformed the input ontology into an RDF graph that reflects the semantics represented in the ontology. Subsequently, we stored this RDF graph in the Neo4j graphical database to ensure efficient and persistent management of RDF data. At the time of the interrogation, we studied the different possible and desired interpretations of the request originally made by the user. We have also proposed to carry out a sort of transformation between the two query languages SPARQL and Cypher, which is specific to Neo4j. This allows us to implement the architecture of our system over a wide variety of BD-RDFs providing their query languages, without changing any of the other components of the system. Finally, we tested and evaluated our tool using different test bases, and it turned out that our tool is comprehensive, effective, and powerful enough.

Multifunction glasses that can monitor your health, let you play video games with your eyes and still work as sunglasses are developed by South Korean scientists. The groundbreaking new wearable tech built at Korea University, Seoul, can provide more advanced personal health data than devices like Fitbits or smart watches. Devices that measure electrical signals from the brain or eyes can help to diagnose conditions like epilepsy and sleep disorders -- as well as in controlling computers. A long-running challenge in measuring these electronic signals, however, has been in developing devices that can maintain the needed steady physical contact between the wearable's sensors and the user's skin. The researchers overcame this issue by integrating soft, conductive electrodes into their glasses that can wirelessly monitor the electrical signals.

'Passive' visual experiences play a key part in our early learning experiences and should be replicated in AI vision systems, according to neuroscientists. Italian researchers argue there are two types of learning – passive and active – and both are crucial in the development of our vision and understanding of the world. Who we become as adults depends on the first years of life from these two types of stimulus – 'passive' observations of the world around us and'active' learning of what we are taught explicitly. In experiments, the scientists demonstrated the importance of the passive experience for the proper functioning of key nerve cells involved in our ability to see. This could lead to direct improvements in new visual rehabilitation therapies or machine learning algorithms employed by artificial vision systems, they claim.

Perhaps the single biggest implication of reopening national economies is that responsibility and thus liability for dealing with the covid-19 pandemic will shift from the public to the private sector. Fortune 500 CEOs and small business owners alike will soon be making decisions that affect the health not only of their business but also their people (employees, contractors, customers, suppliers)--which in turn affects the health of their families, friends, and neighbors. With so much at stake, how should business leaders plan for operating in the post-stay-at-home phase of the recovery? The current crisis is driven by a health problem: we don't yet have a treatment or a vaccine for the novel coronavirus. Managers have little control over that.

The news: The United Nations is endorsing a computer simulation tool that it believes will help governments tackle the world's biggest problems, from gender inequality to climate change. Global challenges: In 2015, UN member states signed up for a set of 17 sustainable-development goals that are due to be reached by 2030. They include things like "zero poverty," "no hunger," and "affordable and clean energy." How could the tool help? Called Policy Priority Inference (PPI), the software uses agent-based modeling to predict what would happen if policymakers spent money on one project rather than another.

There's plenty of news right now about how people are trying to make real money through video games, and not just by trying to get a taste of that Ninja game-streaming fortune. Most recently, people are selling items for hard cash inside the new Animal Crossing: New Horizons. As the coronavirus takes a hammer to the economy and a number of people are at risk of penury, selling goods inside the game seems like a good idea. But while there's plenty of hype about the potential for virtual economies to thrive as the real-world ones collapse, the truth is a little different. If you're unfamiliar, Animal Crossing: New Horizons is a sim game for the Nintendo Switch in which you build a life for yourself in a community of adorable, anthropomorphic animals. You fish, grow fruit, craft tools and furniture while working to improve your island home.