Markov chain Monte Carlo (MCMC) methods are often used in clustering since they guarantee asymptotically exact expectations in the infinite-time limit. In finite time, though, slow mixing often leads to poor performance. Modern computing environments offer massive parallelism, but naive implementations of parallel MCMC can exhibit substantial bias. In MCMC samplers of continuous random variables, Markov chain couplings can overcome bias. But these approaches depend crucially on paired chains meetings after a small number of transitions. We show that straightforward applications of existing coupling ideas to discrete clustering variables fail to meet quickly. This failure arises from the "label-switching problem": semantically equivalent cluster relabelings impede fast meeting of coupled chains. We instead consider chains as exploring the space of partitions rather than partitions' (arbitrary) labelings. Using a metric on the partition space, we formulate a practical algorithm using optimal transport couplings. Our theory confirms our method is accurate and efficient. In experiments ranging from clustering of genes or seeds to graph colorings, we show the benefits of our coupling in the highly parallel, time-limited regime.

The United States Navy announced the launch of a new joint fleet of unmanned drones in the Middle East with allied nations to patrol vast swaths of volatile waters as tensions simmer with Iran. Vice Admiral Brad Cooper, who leads the 5th Fleet, said 100 unmanned drones, both sailing and submersible, would dramatically multiply the surveillance capacities of the US Navy, allowing it to keep a close eye on waters critical to the flow of global oil and shipping. Trade at sea has been targeted in recent years as Tehran's nuclear deal with world powers collapsed. "By using unmanned systems, we can just simply see more. They're high reliability and remove the human factor," Cooper said on the sidelines of a defence exhibition in Abu Dhabi, adding the systems are "the only way to cover on whatever gaps that we have today".

Multi-model ensemble analysis integrates information from multiple climate models into a unified projection. However, existing integration approaches based on model averaging can dilute fine-scale spatial information and incur bias from rescaling low-resolution climate models. We propose a statistical approach, called NN-GPR, using Gaussian process regression (GPR) with an infinitely wide deep neural network based covariance function. NN-GPR requires no assumptions about the relationships between models, no interpolation to a common grid, no stationarity assumptions, and automatically downscales as part of its prediction algorithm. Model experiments show that NN-GPR can be highly skillful at surface temperature and precipitation forecasting by preserving geospatial signals at multiple scales and capturing inter-annual variability. Our projections particularly show improved accuracy and uncertainty quantification skill in regions of high variability, which allows us to cheaply assess tail behavior at a 0.44$^\circ$/50 km spatial resolution without a regional climate model (RCM). Evaluations on reanalysis data and SSP245 forced climate models show that NN-GPR produces similar, overall climatologies to the model ensemble while better capturing fine scale spatial patterns. Finally, we compare NN-GPR's regional predictions against two RCMs and show that NN-GPR can rival the performance of RCMs using only global model data as input.

Neom Tech & Digital Company, a subsidiary of Neom, is building a 3D cognitive digital twin metaverse platform, which aims to enable a "ground-breaking, mixed-reality" model for urban living. Called XVRS, the platform was announced at the Leap22 technology event which has been taking place in Riyadh, Saudi Arabia from 1-3 February. Combining digital and physical architectures with hyper-connected technologies and artificial intelligence (AI), XVRS seeks to seamlessly integrate the virtual and real worlds. Neom is a region in northwest Saudi Arabia on the Red Sea being built from the ground up as a living laboratory. Neom Tech & Digital Company was founded in 2021 as its first subsidiary, charged with helping to create an ecosystem of cognitive technologies to co-invent the future of living.

Tree-based models such as decision trees and random forests (RF) are a cornerstone of modern machine-learning practice. To mitigate overfitting, trees are typically regularized by a variety of techniques that modify their structure (e.g. pruning). We introduce Hierarchical Shrinkage (HS), a post-hoc algorithm that does not modify the tree structure, and instead regularizes the tree by shrinking the prediction over each node towards the sample means of its ancestors. The amount of shrinkage is controlled by a single regularization parameter and the number of data points in each ancestor. Since HS is a post-hoc method, it is extremely fast, compatible with any tree growing algorithm, and can be used synergistically with other regularization techniques. Extensive experiments over a wide variety of real-world datasets show that HS substantially increases the predictive performance of decision trees, even when used in conjunction with other regularization techniques. Moreover, we find that applying HS to each tree in an RF often improves accuracy, as well as its interpretability by simplifying and stabilizing its decision boundaries and SHAP values. We further explain the success of HS in improving prediction performance by showing its equivalence to ridge regression on a (supervised) basis constructed of decision stumps associated with the internal nodes of a tree. All code and models are released in a full-fledged package available on Github (github.com/csinva/imodels)

This paper tackles the problem of missing data imputation for noisy and non-Gaussian data. A classical imputation method, the Expectation Maximization (EM) algorithm for Gaussian mixture models, has shown interesting properties when compared to other popular approaches such as those based on k-nearest neighbors or on multiple imputations by chained equations. However, Gaussian mixture models are known to be not robust to heterogeneous data, which can lead to poor estimation performance when the data is contaminated by outliers or come from a non-Gaussian distributions. To overcome this issue, a new expectation maximization algorithm is investigated for mixtures of elliptical distributions with the nice property of handling potential missing data. The complete-data likelihood associated with mixtures of elliptical distributions is well adapted to the EM framework thanks to its conditional distribution, which is shown to be a Student distribution. Experimental results on synthetic data demonstrate that the proposed algorithm is robust to outliers and can be used with non-Gaussian data. Furthermore, experiments conducted on real-world datasets show that this algorithm is very competitive when compared to other classical imputation methods.

Sertaç Taşdelen, a Turkish entrepreneur and creator of the fortune-telling app Faladdin, does his best to resemble Aladdin's genie. When I recently visited his coworking office in downtown Istanbul, Taşdelen was wearing an electric blue jacket, white trousers, and a fluffy button-down shirt, and his bearded, square-jawed face carried the mischievous smile of the fictional jinn. "Faladdin is my alter ego," Taşdelen said of the psychic he plays in the app. "If I quit business today," he whispered, leaning in, "I'd be a gypsy fortune-teller living in a caravan." In Apple's App Store, Faladdin describes itself as "far beyond a fortune telling app." The description states that it can predict one's destiny "by evaluating a person's past." It does this by bringing the Turkish tradition of coffee fortune-telling into the Internet age.

Group or cluster structure on explanatory variables in machine learning problems is a very general phenomenon, which has attracted broad interest from practitioners and theoreticians alike. In this work we contribute an approach to learning under such group structure, that does not require prior information on the group identities. Our paradigm is motivated by the Laplacian geometry of an underlying network with a related community structure, and proceeds by directly incorporating this into a penalty that is effectively computed via a heat flow-based local network dynamics. In fact, we demonstrate a procedure to construct such a network based on the available data. Notably, we dispense with computationally intensive pre-processing involving clustering of variables, spectral or otherwise. Our technique is underpinned by rigorous theorems that guarantee its effective performance and provide bounds on its sample complexity. In particular, in a wide range of settings, it provably suffices to run the heat flow dynamics for time that is only logarithmic in the problem dimensions. We explore in detail the interfaces of our approach with key statistical physics models in network science, such as the Gaussian Free Field and the Stochastic Block Model. We validate our approach by successful applications to real-world data from a wide array of application domains, including computer science, genetics, climatology and economics. Our work raises the possibility of applying similar diffusion-based techniques to classical learning tasks, exploiting the interplay between geometric, dynamical and stochastic structures underlying the data.

Iran-backed Yemeni fighters launched drone strikes on the United Arab Emirates that caused explosions and a deadly fire outside the capital, Abu Dhabi, ratcheting up security risks in the major oil-exporting region at a critical time. One of the biggest attacks to date on UAE soil ignited a fire at Abu Dhabi's main international airport on Monday and set fuel tanker trucks ablaze in a nearby industrial area. It took place days after Yemen's Houthi fighters warned Abu Dhabi against intensifying its air campaign against them. Crude extended gains to the highest level in seven years on Tuesday after the assaults in the UAE, OPEC's third biggest oil producer. Iran's longtime support of the Houthis means the incidents could roil regional diplomatic efforts to ease frictions and separate talks to restore Tehran's 2015 nuclear deal with world powers.

A suspected drone attack by Yemen's Houthi rebels targeting a key oil facility in Abu Dhabi killed three people and started a separate fire at Abu Dhabi's international airport, police said. Police in the United Arab Emirates identified the dead as two Indian nationals and one Pakistani. "Small flying objects" were found as three petrol tanks exploded in an industrial area and a fire was ignited at the airport, police said, as Houthi rebels announced "military operations" in the UAE. The UAE which had largely scaled down its military presence in Yemen in 2019, continues to hold sway through the Yemeni forces it armed and trained. Drone attacks are a hallmark of the Houthis' assaults on Saudi Arabia, the UAE ally that is leading the coalition fighting for Yemen's government in the grinding civil war.