Testing the significance of a variable or group of variables $X$ for predicting a response $Y$, given additional covariates $Z$, is a ubiquitous task in statistics. A simple but common approach is to specify a linear model, and then test whether the regression coefficient for $X$ is non-zero. However, when the model is misspecified, the test may have poor power, for example when $X$ is involved in complex interactions, or lead to many false rejections. In this work we study the problem of testing the model-free null of conditional mean independence, i.e. that the conditional mean of $Y$ given $X$ and $Z$ does not depend on $X$. We propose a simple and general framework that can leverage flexible nonparametric or machine learning methods, such as additive models or random forests, to yield both robust error control and high power. The procedure involves using these methods to perform regressions, first to estimate a form of projection of $Y$ on $X$ and $Z$ using one half of the data, and then to estimate the expected conditional covariance between this projection and $Y$ on the remaining half of the data. While the approach is general, we show that a version of our procedure using spline regression achieves what we show is the minimax optimal rate in this nonparametric testing problem. Numerical experiments demonstrate the effectiveness of our approach both in terms of maintaining Type I error control, and power, compared to several existing approaches.

By the time California state senators took up a bill Thursday night to grant unemployment benefits to striking workers, labor unions had already scored several monumental wins in the state Legislature. They landed a major deal to raise fast food wages to $20 an hour. They convinced lawmakers to pass a bill requiring driverless trucks to have a human safety driver. They persuaded the Democratic-led Legislature to send Gov. Gavin Newsom a bill giving all workers in California a minimum of five paid sick days -- up from the current requirement of three. So when the time came to vote on allowing striking workers to receive unemployment benefits, an exasperated Republican state senator rose to make the case that businesses wouldn't be able to stay afloat if their employees could get paid while on the picket line.

China-linked hackers are increasingly moving beyond espionage and into the disturbing world of power grid attacks. Threat researchers at security software firm Symantec this week released new evidence that the Chinese hacking group known as APT41 infiltrated the power grid of an Asian nation. Some details of the latest intrusion echo a 2021 attack on India's power grid, suggesting the same hackers are responsible. In Argentina, a scandal is playing out over the use of facial recognition software in Buenos Aires. Despite laws that require authorities to limit searches to known fugitives, an investigation by a judge found that the system was used to look up people not wanted for any crimes.

It has been billed as NASA's'asteroid autumn' and involves a trio of exciting missions that could answer some truly mind-boggling questions. From offering clues to how life on Earth began, to unlocking the secrets of the solar system, key milestones for each voyage are due to play out over the next six weeks. They include one rocket launch, a distant fly-by between Jupiter and Mars, and the recovery of ancient space rocks in the Utah desert that could contain the ingredients for life. Of the three, the lift-off of NASA's Psyche spacecraft probably sounds the most mundane -- but no so fast. That is also a fascinating mission, because it is setting off on a 2.5 billion-mile (4 billion-kilometre) journey to find out once and for all if a metal-rich asteroid really could bring down the world's economy.

Russian President Vladimir Putin and Belarusian President Alexander Lukashenko discussed whether Minsk might join Moscow's efforts to revive an alliance with North Korea, following Kim Jong Un's visit to Russia this week. Finland will prohibit the entry of vehicles with Russian licence plates as of midnight on Saturday, following Estonia, Latvia and Lithuania, which have also recently barred all Russian-registered cars from crossing their borders. Romania has imposed additional flight restrictions in parts of its air space along the border with Ukraine amid a surge in Russian drone attacks on nearby Ukrainian Danube river ports. Russian President Vladimir Putin and Belarusian President Alexander Lukashenko discussed whether Minsk might join Moscow's efforts to revive an alliance with North Korea, following Kim Jong Un's visit to Russia this week. Finland will prohibit the entry of vehicles with Russian licence plates as of midnight on Saturday, following Estonia, Latvia and Lithuania, which have also recently barred all Russian-registered cars from crossing their borders.

Organizations of all kinds, whether public or private, profit-driven or non-profit, and across various industries and sectors, rely on dashboards for effective data visualization. However, the reliability and efficacy of these dashboards rely on the quality of the visual and data they present. Studies show that less than a quarter of dashboards provide information about their sources, which is just one of the expected metadata when provenance is seriously considered. Provenance is a record that describes people, organizations, entities, and activities that had a role in the production, influence, or delivery of a piece of data or an object. This paper aims to provide a provenance representation model, that entitles standardization, modeling, generation, capture, and visualization, specifically designed for dashboards and its visual and data components. The proposed model will offer a comprehensive set of essential provenance metadata that enables users to evaluate the quality, consistency, and reliability of the information presented on dashboards. This will allow a clear and precise understanding of the context in which a specific dashboard was developed, ultimately leading to better decision-making.

In the contemporary digital age, Quantum Computing and Artificial Intelligence (AI) convergence is reshaping the cyber landscape, introducing unprecedented opportunities and potential vulnerabilities.This research, conducted over five years, delves into the cybersecurity implications of this convergence, with a particular focus on AI/Natural Language Processing (NLP) models and quantum cryptographic protocols, notably the BB84 method and specific NIST-approved algorithms. Utilising Python and C++ as primary computational tools, the study employs a "red teaming" approach, simulating potential cyber-attacks to assess the robustness of quantum security measures. Preliminary research over 12 months laid the groundwork, which this study seeks to expand upon, aiming to translate theoretical insights into actionable, real-world cybersecurity solutions. Located at the University of Oxford's technology precinct, the research benefits from state-of-the-art infrastructure and a rich collaborative environment. The study's overarching goal is to ensure that as the digital world transitions to quantum-enhanced operations, it remains resilient against AI-driven cyber threats. The research aims to foster a safer, quantum-ready digital future through iterative testing, feedback integration, and continuous improvement. The findings are intended for broad dissemination, ensuring that the knowledge benefits academia and the global community, emphasising the responsible and secure harnessing of quantum technology.

The concepts of conditional mutual information (CMI) and normalized conditional mutual information (NCMI) are introduced to measure the concentration and separation performance of a classification deep neural network (DNN) in the output probability distribution space of the DNN, where CMI and the ratio between CMI and NCMI represent the intra-class concentration and inter-class separation of the DNN, respectively. By using NCMI to evaluate popular DNNs pretrained over ImageNet in the literature, it is shown that their validation accuracies over ImageNet validation data set are more or less inversely proportional to their NCMI values. Based on this observation, the standard deep learning (DL) framework is further modified to minimize the standard cross entropy function subject to an NCMI constraint, yielding CMI constrained deep learning (CMIC-DL). A novel alternating learning algorithm is proposed to solve such a constrained optimization problem. Extensive experiment results show that DNNs trained within CMIC-DL outperform the state-of-the-art models trained within the standard DL and other loss functions in the literature in terms of both accuracy and robustness against adversarial attacks. In addition, visualizing the evolution of learning process through the lens of CMI and NCMI is also advocated.

Large language models are quickly gaining momentum, yet are found to demonstrate gender bias in their responses. In this paper, we conducted a content analysis of social media discussions to gauge public perceptions of gender bias in LLMs which are trained in different cultural contexts, i.e., ChatGPT, a US-based LLM, or Ernie, a China-based LLM. People shared both observations of gender bias in their personal use and scientific findings about gender bias in LLMs. A difference between the two LLMs was seen -- ChatGPT was more often found to carry implicit gender bias, e.g., associating men and women with different profession titles, while explicit gender bias was found in Ernie's responses, e.g., overly promoting women's pursuit of marriage over career. Based on the findings, we reflect on the impact of culture on gender bias and propose governance recommendations to regulate gender bias in LLMs.

Node features bolster graph-based learning when exploited jointly with network structure. However, a lack of nodal attributes is prevalent in graph data. We present a framework to recover completely missing node features for a set of graphs, where we only know the signals of a subset of graphs. Our approach incorporates prior information from both graph topology and existing nodal values. We demonstrate an example implementation of our framework where we assume that node features depend on local graph structure. Missing nodal values are estimated by aggregating known features from the most similar nodes. Similarity is measured through a node embedding space that preserves local topological features, which we train using a Graph AutoEncoder. We empirically show not only the accuracy of our feature estimation approach but also its value for downstream graph classification. Our success embarks on and implies the need to emphasize the relationship between node features and graph structure in graph-based learning.