Signals, in order.

The event titled "From Patterns to Principles: Machine Learning-Informed Theory Formation Methods for the Social Sciences" is a research gathering in Zurich, Switzerland. It is led by Prof. Dr. Caspar van Lissa of Tilburg University and will take place at AFL, Affolternstrasse 56, room F-121.

Source: UZH Events

X-ray photoelectron spectroscopy (XPS) is used to probe chemical states in battery materials, but complex microstructures hinder quantitative interpretation. The study introduces a computational framework that combines machine‑learning prediction of core‑level binding energies with grand‑canonical Monte Carlo and molecular dynamics simulations to systematically sample lithiation states and local atomic environments in silicon anodes. The resulting stoichiometry maps reproduce key spectroscopic trends observed in operando and ex situ XPS, including Si 2p signatures of crystal‑to‑amorphous diso…

Source: OpenAlex

On 16 October 2026, ETH Zurich inaugurated the ETH Swiss GeoLab in Root near Lucerne, a centre focused on early detection of natural hazards and optimisation of agriculture and forestry through satellite and drone data. The lab will develop an AI‑based geoinformation platform that integrates satellite imagery, drone footage, monitoring station data, and other spatial information, and will collaborate with the Swiss National Supercomputing Centre to train AI models on large Earth observation datasets. Initial tests include drone‑mounted radar systems in the Lötschental valley to detect surface…

Source: ETH Zurich News

Yoon Seok Kim, an Assistant Professor of Life Science Engineering at EPFL, has collaborated with three Nobel laureates, including Karl Deisseroth, whose 2026 Nobel Prize in Physiology or Medicine recognized the discovery of light‑gated ion channels that underpin optogenetics. Kim contributed to the first crystal structure of channelrhodopsins, enabling the design of improved optogenetic tools, and now leads a laboratory at EPFL that studies mechanosensitive proteins and cancer neuroscience. His work continues to build on the foundations of optogenetics and protein engineering developed during…

Source: EPFL News

ETH Zurich’s new Master’s in Space Systems has rapidly attracted high demand, growing from 27 to 47 students in two years and producing its first graduates who are now employed by aerospace companies in Germany and the United States. The program combines technical coursework in launch vehicles, satellites, and planetary sciences with interdisciplinary team projects that involve industry partners, research institutions, and start‑ups, fostering a holistic view of complex space systems. The curriculum is designed to support students in developing startup ideas and to integrate applied sciences …

Source: ETH Zurich News

StellarSpectraObservationFitting.jl (SSOF) is a Julia package that measures Doppler shifts and creates data-driven models for time-variable spectral features, including both telluric absorption and stellar variability, while accounting for the wavelength-dependent instrumental line spread function. It employs fast, physically motivated Gaussian Process regularization and demonstrates state-of-the-art performance on NEID spectrometer data, achieving 2–6× smaller photon-limited errors compared to the NEID CCF-based pipeline and matching the performance of the SERVAL template-based pipeline usin…

Source: OpenAlex

Herpes zoster is typically identified by painful dermatomal eruptions, but motor complications such as abdominal pseudohernia are often overlooked. In this case, a 66‑year‑old man developed a painful unilateral abdominal bulge four days before the characteristic rash, with imaging showing no fascial defect and laboratory tests remaining normal. The bulge resolved with conservative management, highlighting the need to consider herpes zoster when a new segmental abdominal bulge appears, even before skin lesions manifest.

Source: OpenAlex

ETH Zurich researchers have developed a method to use bacteria to bind sand into solid architectural elements, creating a low‑energy alternative to cement. By combining microbiologically induced calcite precipitation with 3D printing and robotic spraying, they can produce free‑form structures and surfaces that harden at room temperature. The process, which takes two to five days, relies on bacteria breaking down urea to form calcium carbonate that interlocks sand grains.

Source: ETH Zurich News

The study presents a physics-informed machine learning framework that analyzes real-time sensor data from over 25 wells to detect and predict water conformance issues in hydrocarbon production. By distinguishing between near-wellbore and reservoir-scale problems, the system forecasts water breakthrough events and guides targeted shutoff treatments, potentially improving recovery and reducing OPEX. The approach transforms produced water management from reactive to data-driven decision support.

Source: OpenAlex

The paper presents a symmetry-adapted machine learning framework for predicting tensorial properties of atomistic systems. It demonstrates improved accuracy and efficiency compared to conventional methods. The research originates from Switzerland.

Source: opendata.swiss