The building that houses Recruit Holdings' headquarters in Tokyo. Investors have shown confidence in Recruit's HR Technology unit, where Indeed is leveraging AI to improve matching and raising average revenue from each job posting even as hiring demand remains soft. Recruit Holdings shares climbed the most on record after the Japanese owner of Indeed.com The stock jumped as much as 19%, its biggest intraday increase since the company went public in 2014, even as the Topix index fell, after issuing an outlook for ¥787 billion ($5 billion) in operating profit on ¥4 trillion in sales for the fiscal year to March 2027. That exceeds analysts' average projection for ¥723 billion and ¥3.9 trillion, respectively.

Shares are up more than 300% this year for the Tokyo-based company. Kioxia Holdings' shares were untraded in a glut of buy orders Monday morning after the supplier of storage for artificial intelligence data centers reported soaring profit and gave an outlook that trounced expectations. The Tokyo-based company said it expects to earn an operating profit of ¥1.3 trillion ($8.2 billion) during the June quarter, above the record profit it earned for the full year ended March. Its quarterly profit also surged past expectations, surpassing that of Toyota's, making Kioxia one of Japan's most profitable businesses. Kioxia's shares are up more than 300% this year.

The killing of Abu-Bilal al-Minuki, the second-in-command of ISIL (ISIS), by United States and Nigerian forces marks a notable achievement for "counterterrorism". Yet for analysts observing the Lake Chad Basin, it highlights how persistent and complex insecurity in the region has become. Al-Minuki, a Nigerian national from Borno State, was operating out of a compound near Lake Chad, at the centre of one of the world's most active armed group theatres. Perhaps equally significant is the parallel resurgence of Boko Haram, which quietly rebuilt itself while security agencies primarily focused on the more dominant ISWAP. "While regional forces focused on countering ISWAP's threats, partly due to the group's advanced drone capabilities, Boko Haram appears to have taken advantage of the relative attention on its rival to regroup," Nimi Princewill, a security expert in the Sahel, told Al Jazeera.

Cuba says U.S. fabricating pretext for conflict after report on drone purchase Cuba's Foreign Minister Bruno Rodriguez speaks during a news conference in Havana in October 2025. HAVANA - Cuban Foreign Minister Bruno Rodriguez accused the U.S. on Sunday of fabricating a fraudulent case to justify economic sanctions and potential military intervention. The minister's comments followed a report by Axios the same day citing classified intelligence, which said Cuba had acquired more than 300 military drones. Cuba neither threatens nor desires war, Rodriguez said in a post on social media, adding that the country prepares itself to confront external aggression in the exercise of the right to legitimate self-defense recognized by the U.N. Charter. In a time of both misinformation and too much information, quality journalism is more crucial than ever.

U.S. and Iran far from Hormuz deal as Trump says clock is ticking A ship remains anchored in the Strait of Hormuz near Larak Island, Iran, on Saturday. The U.S. and Iran remained far apart Sunday on a deal to end weeks of war and reopen the crucial Strait of Hormuz, as a drone attack sparked a fire at a United Arab Emirates nuclear plant, spotlighting the risks of a fragile ceasefire. U.S. President Donald Trump made clear his patience is wearing thin, posting on social media Sunday that For Iran, the Clock is Ticking, and they better get moving, FAST, or there won't be anything left of them. TIME IS OF THE ESSENCE!" Iran's semiofficial Fars news agency said the U.S. had set five main conditions for a peace deal, including the removal of uranium used by Iran's nuclear program to the U.S.; no U.S. reparations to Tehran and the unfreezing of less than a quarter of Iran's suspended assets. Fars didn't give a source for the information, and the U.S. hasn't publicly commented on such stipulations.

Could the war trigger a hunger crisis? How well do you know Iran? This video may contain light patterns or images that could trigger seizures or cause discomfort for people with visual sensitivities. US President Donald Trump warns Iran that the "clock is ticking" for a peace deal to be reached with Washington. Saudi Arabia says it intercepted three drones, as the UAE reported a separate drone strike near its Barakah nuclear power plant that sparked a fire.

This paper proposes a deep reinforcement learning (DRL)-based event-triggered controller design for networked artificial pancreas (AP) systems. Although existing DRL-based AP controllers typically assume periodic control updates, networked control systems (NCSs) require a reduction in communication frequency to achieve energy-efficient operation, which is directly tied to control updates. However, jointly learning both insulin dosing and update timing significantly increases the complexity of the learning problem. To alleviate this complexity, we develop a practical DRL-based controller design that avoids explicitly learning update timing by introducing a rule-based criterion defined by changes in blood glucose. As a result, decision-making occurs at irregular intervals, and the problem is naturally formulated as a semi-Markov decision process (SMDP), for which we extend a standard DRL algorithm. Numerical experiments demonstrate that the proposed method improves communication efficiency while maintaining control performance.

Coherent Point Drift (CPD) is a representative probabilistic framework for unsupervised non-rigid point set registration. Its standard non-rigid M-step, however, relies on a point-indexed Gaussian-kernel system whose size grows with the number of moving points, making deformation estimation computationally heavy for large point sets and difficult to control in complexity during registration. To address these limitations, we propose Analytic-CPD, a new unsupervised non-rigid registration framework that gives CPD a structured analytic reformulation. Analytic-CPD preserves the CPD posterior correspondence layer, but lifts the M-step from point-indexed kernel displacement estimation to structured analytic mapping estimation. By coupling the Gaussian-mixture posterior mechanism of CPD with Structured Analytic Mappings (SAM), the method obtains a deformation model whose coefficient dimension is governed by the ambient dimension and analytic order rather than by the number of moving points. More importantly, deformation estimation is organized over an interpretable hierarchy of analytic function spaces, so the analytic order can be increased progressively as posterior correspondences become more reliable. We implement this idea through an increasing-degree continuation strategy with decreasing stage lengths: low-order analytic maps first stabilize the posterior correspondence structure, while higher-order modes later refine nonlinear residual deformation. Experiments on controlled model-matched, smooth model-mismatch, and registered human-shape data demonstrate the effectiveness and favorable accuracy--efficiency performance of Analytic-CPD.

By far the most common way to estimate an expected loss in machine learning is to draw samples, compute the loss on each one, and take the empirical average. However, sampling is not necessarily optimal. Given an MLP at initialization, we show how to estimate its expected output over Gaussian inputs without running samples through the network at all. Instead, we produce approximate representations of the distributions of activations at each layer, leveraging tools such as cumulants and Hermite expansions. We show both theoretically and empirically that for sufficiently wide networks, our estimator achieves a target mean squared error using substantially fewer FLOPs than Monte Carlo sampling. We find moreover that our methods perform particularly well at estimating the probabilities of rare events, and additionally demonstrate how they can be used for model training. Together, these findings suggest a path to producing models with a greatly reduced probability of catastrophic tail risks.

Deep generative models provide flexible frameworks for modeling complex, structured data such as images, videos, 3D objects, and texts. However, when applied to sequences of human skeletons, standard variational autoencoders (VAEs) often allocate substantial capacity to nuisance factors-such as camera orientation, subject scale, viewpoint, and execution speed-rather than the intrinsic geometry of shapes and their motion. We propose the Elastic Shape - Variational Autoencoder (ES-VAE), a geometry-aware generative model for skeletal trajectories that leverages the transported square-root velocity field (TSRVF) representation on Kendall's shape manifold. This representation inherently removes rigid translations, rotations, and global scaling of shapes, and temporal rate variability of sequences, isolating the underlying shape dynamics. The ES-VAE encoder maps skeletal sequences to a low-dimensional latent space incorporating the Riemannian logarithm map, while the decoder reconstructs sequences using the corresponding exponential map. We demonstrate the effectiveness of ES-VAE on two datasets. First, we analyze skeletal gait cycles to predict clinical mobility scores and classify subjects into healthy and post-stroke groups. Second, we evaluate action recognition on the NTU RGB+D dataset. Across both settings, ES-VAE consistently outperforms standard VAEs and a range of sequence modeling baselines, including temporal convolutional networks, transformers, and graph convolutional networks. More broadly, ES-VAE provides a principled framework for learning generative models of longitudinal data on pose shape manifolds, offering improved latent representation and downstream performance compared to existing deep learning approaches.