ForecastingWhen quantum meets AI: NUS brings together international experts to turn frontier technologies into real-world solutionsNational University of Singapore — News · 5hModellingHigher-Order Approximation of Exit Functionals in Sampling-Based Stochastic Model Predictive ControlarXiv — Systems and Control (eess.SY) · yesterdayDigital twinPhysics-Informed Neural Networks for Climate-Aware Digital TwinsarXiv API — полнотекстовый поиск "digital twin" · yesterdayModellingРоссийские компании могут оценить готовность собственной инфраструктуры к агентной разработкеCNews · 22hModellingGigaCode Desktop — мультиагентное приложение Сбербанка для управления жизненным циклом разработки продуктов — стало доступно бизнесуCNews · 23hModellingАгентный AI и мобильный гейминг нового уровня: Qualcomm представила флагманские 2-нм чипсеты Snapdragon 8 Elite Extreme Gen 6 и 8 Elite Gen 6Gagadget · 23hForecastingCollective Tube MPC With Distribution-Free Joint Safety CertificatesarXiv — Systems and Control (eess.SY) · yesterdayForecastingSampling-based MPC Using Trust RegionsarXiv — Systems and Control (eess.SY) · yesterdayForecastingLinearly Scalable Nonlinear MPC with Adaptive HorizonsarXiv — Systems and Control (eess.SY) · yesterdayModellingPhysics-Informed Neural Network Surrogates with Polynomial Chaos-Based Uncertainty Propagation for Stochastic Model Predictive ControlarXiv — Systems and Control (eess.SY) · 2dModellingEnergy-based Regularization for Learning Residual Dynamics in Neural MPC for Omnidirectional Aerial RobotsarXiv — Systems and Control (eess.SY) · 2dModellingSocioVerse2: A Longitudinal Dynamic Social Simulation Framework under a Human-AI Co-evolutionary ParadigmarXiv — Computers and Society (cs.CY) · 2d
Function+Data Flow DSL Makes Building AI Pipelines for Digital Twins Easier
US & West ·
The paper introduces Function+Data Flow (FDF), a visual domain-specific language for specifying machine learning pipelines used to build real-time digital twins. Implemented in DesCartes Builder, an integrated modeling environment, FDF aims to make AI-based digital twin development more accessible and reliable, as evaluated in a user study.
Why it matters
Visual DSLs for AI pipelines enhance reproducibility and manageability of digital twin engineering, relevant for modeling and control of complex systems.
The National University of Singapore (NUS) and the Society of Algorithmic Intelligence (SoAI) is hosting IntelligenceX 2026: Global Quantum × AI Frontier, from 24 to 26 September 2026 at NUS University Town, focused on advances at the intersection of quantum computing and artificial intelligence…
This study evaluates a physics-informed neural network for potential-temperature forecasting under incomplete thermal observations, constrained by a pressure-coordinate advection-source equation and a diabatic-source closure frozen after the preceding 12 hours. Validation uses hourly ERA5 reanalysis at three pressure levels for one-, two- and three-hour horizons against persistence, local-trend, and two neural baselines.
To reduce computational costs of obtaining channel state information in 5G NR networks, three approaches are studied within a calibrated 3D digital twin: generalizing the channel twin, enhancing performance with a neural receiver, and a data-driven interpolation framework for scalability. The goal is to meet strict microsecond latency budgets of OFDM symbols.