New AI method for simulating high-speed flows

Gompertz Linear Unit (GoLU): a new era in activation functions Activation functions are the heartbeat of neural networks, they decide how neurons “fire” and what patterns models learn. From ReLU and GELU to Swish and Mish, each innovation has refined how information flows through deep architectures. This year, Indrashis Das, Mahmoud Safari, Steven Adriaensen, and Frank Hutter introduced GoLU from The University of Freiburg and Prior Labs. Definition GoLU(x) = x · e^(-e^(-x)) A smooth, asymmetric activation that stabilizes learning while maintaining gradient flow. Why it matters • Right-leaning asymmetry reduces activation variance and smooths training • Flatter loss landscapes and more stable optimization • Broader weight distributions that capture richer features • Strong results across vision, language, and diffusion benchmarks, often outperforming ReLU, GELU, and Swish Code: https://lnkd.in/dzkDgyNx Paper: https://lnkd.in/dsmX_8pE #DeepLearning #NeuralNetworks #ActivationFunctions #GoLU #MachineLearning #AIResearch

📢 #Editor_Choices' paper of #Water journal 📄 Graph Neural Networks for Sensor Placement: A Proof of Concept towards a Digital Twin of Water Distribution Systems ✍️ Andrea Menapace, Ariele Zanfei, Manuel Herrera and Bruno Brentan Find out more 👉 https://brnw.ch/21wWFXH

🔥 Hot off the oven from our lab! Our latest paper is now published in Water Resources Research! 🔥 “Probabilistic Physics-Guided Deep Neural Networks With Recurrence and Attention Mechanisms for Interpretable Daily Streamflow Simulation” explores the exciting frontier where hydrologic science meets advanced AI. By integrating Temporal Fusion Transformers (TFT) and DeepAR with physical catchment attributes, we developed models that deliver more accurate, interpretable, and uncertainty-aware streamflow predictions across U.S. basins. 💡 This is a major step toward enhancing the interpretability, uncertainty, and explainability of AI applications in water resources — paving the way for more transparent, trustworthy, and science-grounded AI in environmental decision-making.💧 Read the full paper here: https://lnkd.in/eVB3UBuh #PhysicsInformedAI #ExplainableAI #Interpretability #Uncertainty #WaterResources

This “Chemometrics in Spectroscopy” column traces the historical and technical development of machine learning methods, emphasizing their application in calibrating spectrophotometers for predicting measured sample chemical or physical properties—particularly in near-infrared (NIR), infrared (IR), Raman, and atomic spectroscopy—and explores how AI and deep learning are reshaping the spectroscopic landscape. Read more here: https://hubs.li/Q03P59Mg0

3D Full-body Pose Estimation of Construction Machines Using Deep Neural Network and Stereo Vision Paper ID: 280 – ISARC 2025 by Han Luo IAARC https://lnkd.in/dwf9mmEf

🚀 Quantum-enhanced Computer Vision (QeCV): The Future of Visual AI A new survey paper dives into the emerging field of Quantum-enhanced Computer Vision—at the crossroads of computer vision, machine learning, optimization theory, and quantum computing. QeCV promises breakthroughs powered by quantum mechanics, making computations and solutions possible that classical algorithms struggle with, especially for time-intensive and complex problems. Key insights: Quantum computing can drastically improve time scalability for visual processing problems. Parametrized quantum circuits may become real alternatives to classical neural networks in CV applications. This survey delivers a comprehensive introduction to QeCV, details hardware-compatible algorithms, and highlights two major paradigms: gate-based quantum computing and quantum annealing. Awesome work by Natacha Kuete Meli et al! In our Vision AI weekly newsletter, we cover the latest updates in the Vision AI space. Interested to know more? Link below 👇 #QuantumComputing #ComputerVision #AI #QeCV #TechInnovation #arXiv #MachineLearning #VisionAI

Check out the talk I gave at the Perimeter Institute for Theoretical Physics, called "The Lay of the Land in AI". In 1 hour, we managed to cover lots of topics! - Predictive ML: Regression, classification, neural networks, the Universal Approximation Theorem, CNNs, RNNs - Generative AI: Autoencoders, GANs, stable diffusion (CLIP), embeddings, transformers, attention mechanism. Thanks to Christine Muschik and Achim Kempf for the invitation! https://pirsa.org/25090052

📢 New Review Article Published! “A Review: Remote Sensing Image Object Detection Algorithm Based on Deep Learning” This paper surveys the progress in object detection for remote sensing imagery, focusing on deep learning–based methods including the YOLO series, SSD, candidate region (two-stage) approaches, and Transformer-based detectors. 🔗 Read the full open-access review here: https://lnkd.in/dgd5BrN9 #RemoteSensing #DeepLearning #ObjectDetection #YOLO #Transformers #ComputerVision #SurveyArticle #MDPI

Interesting research done. Analysis of Instantaneous Energy Consumption and Recuperation in Electric Buses During SORT Tests Using Linear and Neural Network Models https://lnkd.in/ewe-YeUE

🚀 Revolutionizing Material Science with AI Innovation Researchers at MIT have introduced SCIGEN, a cutting-edge tool that uses Generative AI to design breakthrough materials with quantum properties — accelerating the future of materials discovery and innovation. This marks a major step toward AI-driven scientific research, where machine learning doesn’t just predict but creates. 🔬✨ 💡 AI meets Quantum Science. Innovation meets Discovery. #AIInnovation #MaterialScience #MITResearch #QuantumTechnology #AIDrivenDiscovery #FutureOfScience #TechBreakthrough #GenerativeAI #DeepTech #ArtificialIntelligence #Innovation #ResearchAndDevelopment #TechNews #DigitalTransformation