We consider the problem of conditional independence (CI) testing and adopt a kernel-based approach. Kernel-based CI tests embed variables in reproducing kernel Hilbert spaces, regress their embeddings on the conditioning variables, and test the resulting residuals for marginal independence. This approach yields tests that are sensitive to a broad range of conditional dependencies. Existing methods, however, rely heavily on kernel ridge regression, which is computationally expensive when properly tuned and yields poorly calibrated tests when left untuned, which limits their practical usefulness. We propose the Generalised Kernel Covariance Measure (GKCM), a regression-model-agnostic kernel-based CI test that accommodates a broad class of regression estimators. Building on the Generalised Hilbertian Covariance Measure framework (Lundborg et al., 2022), we characterise conditions under which GKCM satisfies uniform asymptotic level guarantees. In simulations, GKCM paired with tree-based regression models frequently outperforms state-of-the-art CI tests across a diverse range of data-generating processes, achieving better type I error control and competitive or superior power.

Decision-making problems often feature uncertainty stemming from heterogeneous and context-dependent human preferences. To address this, we propose a sequential learning-and-optimization pipeline to learn preference distributions and leverage them to solve downstream problems, for example risk-averse formulations. We focus on human choice settings that can be formulated as (integer) linear programs. In such settings, existing inverse optimization and choice modelling methods infer preferences from observed choices but typically produce point estimates or fail to capture contextual shifts, making them unsuitable for risk-averse decision-making. Using a bounded-variance score function gradient estimator, we train a predictive model mapping contextual features to a rich class of parameterizable distributions. This approach yields a maximum likelihood estimate. The model generates scenarios for unseen contexts in the subsequent optimization phase. In a synthetic ridesharing environment, our approach reduces average post-decision surprise by up to 114$\times$ compared to a risk-neutral approach with perfect predictions and up to 25$\times$ compared to leading risk-averse baselines.

The app reads your email inbox and your meeting calendar, then gives you a short audio summary. It can help you spend less time scrolling, but of course, there are privacy drawbacks to consider.

A new book from Rebecca Solnit, promising to bring us hope in these "difficult times", is among our pick of popular science titles out this month - along with a guide on how to talk to AI, and a look at modern warfare March, in the northern hemisphere anyway, is about venturing out for some much-needed vitamin D and dodging showers. Forget that - just head for a decent café where you can delve into the marvellous science books we've got waiting for you. This month you can explore how animals shaped our world, how to spot liars from their language, what forest trees can tell us - and flowers as revolutionaries. There is some stronger stuff too, if you are in the mood: try AI in the hands of the US military, or a deep cultural look at how our world has changed beyond recognition. Whatever your choice, it's all guaranteed to enrich the inner you.

Google is updating its AI search results to display links more prominently after facing criticism from publishers about reduced website traffic. PCWorld reports that the changes include clearer link icons on mobile and desktop, plus new pop-up windows showing source lists with article descriptions and images. These improvements aim to help users better discover web content while addressing concerns about AI Overviews potentially harming publisher visibility. Google is updating how links are displayed in Search's AI results, making them clearer and more prominent, reports The Verge . According to a social media post by Robby Stein, Vice President of Google Search, internal testing shows that the new AI results interface makes it easier for users to find content on the web. New on Search: In AI Overviews and AI Mode, groups of links will automatically appear in a pop-up as you hover over them on desktop, so you can jump right into a website to learn more.

To this end, we present a formulation of synthesis planning with starting material constraints.

However, progress is limited by a lack of standardized datasets and comparison metrics, causing methods to be compared in an ad hoc manner.

The optical transparent layers, which are trained with an online training approach, backpropagating the error to the analytical model of the system, are passive and kept the same across different steps of denoising. Hence this method enables high-speed image generation with minimal power consumption, benefiting from the bandwidth and energy efficiency of optical informationprocessing.