Centrum Wiskunde & Informatica’s Post

This is the story of GedankenNet, a form of AI that only knows the wave equation to teach itself how to reconstruct holograms using physics. The first time it comes across an experimental hologram, it can reconstruct microscopic images of samples better than supervised learning-based models. "Self-supervised learning of hologram reconstruction using physics consistency" at Nature Machine Intelligence: https://lnkd.in/gmvYeQ8C UCLA Henry Samueli School of Engineering and Applied Science ECE Department California NanoSystems Institute at UCLA #PhysicsDrivenLearning

AI Revolutionizes Physics: Data to Discovery --------------------- 🌌 Imagine AI as the New Telescope in Physics Discoveries! The fusion of Artificial Intelligence \

👉🏻 MICROSOFT WEBINAR FOR RESEARCHERS on June 19 AI for Research: How to Harness the Power of AI and HPC for Scientific Discovery ❗❕ 📆 Wednesday, June 19th, 10:00 -11:30AM (GMT 2) 🔔An event to explore most pressing questions facing our research communities, to learn about the recent research advances, and to discuss bold ideas on how to ensure new technologies and AI could have the broadest possible benefit for the humanity. 🔛 If you work on multiscale modeling, computational methods in biology, chemistry, physics medicine, engineering and/or use HPC and AI in your daily work, this event is for you. 🔗Registration 👇🏻 https://lnkd.in/dq8mm7Wi

Is uncertainties handling the ultimate roadblock for a wider adoption of AI in science ? A workshop on AI and the Uncertainty Challenge in Fundamental Physics, 27-Nov-1-Dec with sessions on Uncertainty Quantification, Simulation Based Inference and more, https://lnkd.in/e-G7nPXd At SCAI, Jussieu, Paris and at Institut Pascal Paris-Saclay Also a Fair-Universe hackathon, a prototype challenge (aiming to be a full blown Neurips 2024 competitions) were participants should estimate the Higgs boson contribution, taking into account input uncertainties but also reporting a Confidence Interval instead of a single measurement. A first in scientific competitions !

In the latest development in the world of physics, two independent teams have made a breakthrough discovery. According to Physics World, they have found that reinforcement learning, a form of machine learning, can help cold-atom systems handle disruptions. This is a promising development for the field, as it has the potential to make cold-atom experiments much easier to handle. To learn more about this exciting discovery, check out the article below. Source: https://lnkd.in/dQAr96pD

Machine Learning Approach for Event Position Reconstruction in the DEAP-3600 Dark Matter Search Experiment | Communication by DEAP Collaboration https://lnkd.in/gwu242he MDPI; DEAP Collaboration #darkmatter #liquidargon #detector #machinelearning #neuralnetwork #physics This article belongs to the Special Issue: From Heavy Ions to Astroparticle Physics https://lnkd.in/guKj_d_M #Abstract In addition to classical analytical data processing methods, machine learning methods are widely used for data analysis in elementary particle physics. Most often, such techniques are used to identify a particular class of events (the classification problem) or to predict a certain event parameter (the regression problem). Here, we present the result of using a machine learning model to solve the regression problem of event position reconstruction in the DEAP-3600 dark matter search detector. A neural network was used as a machine learning model. Improving the position resolution will improve the reduction in background events, while increasing the signal acceptance for weakly interacting massive particles.

How GPT-4o Will Change Education📚 A student shares their iPad screen with the new GPT-4o, and the AI speaks with him and helps him learn in real time. 🌎 Read more: https://lnkd.in/eiW7T3JG 📥 Latest newsletter: https://lnkd.in/d7B7fqA #engineeredmind #science #technology #engineering #mechanicalengineering #physics #mathematics #cfd #simulation

💥 Tutoring and interaction with #ai 💥 Education system turning towards humenless tutoring and mentoring #gpt4o #ai #ml #tutoring #education #knowledge #fundamentals #futureofeducation #shappingfuture

Physics meets machine learning! In this original article, Liqun Shan et al. propose a novel method to solve the Buckley-Leverett partial differential equations using a physics-informed neural network with long short-term memory & attention mechanism. 'Physics-informed machine learning for solving partial differential equations in porous media' features in Advances in Geo-Energy Research. Check it out on #SciOpen: https://lnkd.in/e4jpQ_jK #physics #machinelearning #neuralnetworks #porousmedia #GeoEnergy #Research