Latest Trends in Simulation Techniques

Quantum Simulator Merges Digital and Analog Modes for Unprecedented Precision in Physics Calculations Scientists from Google and universities across five countries, in collaboration with theoretical physicists Andreas Läuchli and Andreas Elben at PSI, have developed a groundbreaking digital-analog quantum simulator capable of calculating complex physical processes with unprecedented precision. Their research, published in Nature on February 5, brings us closer to realizing Richard Feynman’s 1982 vision of quantum simulation as a superior alternative to classical computing for physics problems. Key Advancements • Overcoming Classical Computing Limitations: • Even the fastest supercomputers struggle with simulating quantum processes, such as how cold milk disperses in hot coffee. • Quantum simulators, unlike classical computers, can efficiently model quantum behaviors by replicating the underlying physics within their own quantum states. • Hybrid Digital-Analog Approach: • The new simulator combines digital quantum gates with high-fidelity analog evolution, allowing it to simulate a broader range of physical systems than purely digital or purely analog approaches. • This flexibility enables simulations across solid-state physics, condensed matter, and even astrophysical processes. • Scalability and Precision: • Unlike previous quantum simulators, this design is highly scalable, making it applicable to a wide range of scientific problems with higher accuracy than classical models. Why This Matters • Accelerating Scientific Discoveries: The simulator can model real-world physical systems more efficiently, impacting materials science, quantum chemistry, and fundamental physics. • Bridging the Gap Between Theory and Experimentation: The ability to simulate quantum interactions with extreme accuracy allows researchers to test theoretical models that were previously impossible to verify. • Potential for a Quantum Computing Breakthrough: This hybrid approach demonstrates the power of quantum simulation, potentially leading to practical, scalable quantum computers capable of solving real-world problems. What’s Next? • Expanding the Simulator’s Applications: Researchers will explore how this hybrid digital-analog approach can be applied to more complex quantum systems. • Scaling Up Quantum Simulations: Larger quantum processors will be tested to further push the limits of computational physics. • Collaboration with Industry & Research Institutions: Google and academic institutions are likely to integrate this technology into broader quantum computing efforts, enhancing its practical applications. This milestone in quantum simulation represents a major step toward realizing quantum computing’s potential, proving that hybrid quantum approaches may be the key to unlocking the next era of scientific computing.

Virtual simulations from game tech will increasingly be used for real-world applications Last week we at Andreessen Horowitz did a round up of the Big Ideas for 2025 Here was mine: Traditionally, games have been virtual world simulations designed for fun. Now gaming technology is extending beyond entertainment to transform how businesses operate. While gaming has long pioneered breakthrough technologies — from Nvidia’s graphics to Unreal Engine’s real-time 3D rendering — these tools are now solving critical business challenges. Consider Applied Intuition, a company built on Unreal Engine, which creates virtual simulations to train and test autonomous vehicles. Three forces are accelerating this shift: generative AI is slashing the cost of virtual content creation; advanced 3D capture technologies are digitizing real-world environments (aka digital twins); and next-generation XR devices are making immersive experiences practical for workers. The applications are already here: Anduril Industries leverages game engines for defense simulations; Tesla  creates virtual worlds for autonomous systems; BMW is incorporating AR in future heads-up display systems; Matterport revolutionizes real estate with virtual walkthroughs; Traverse3D helps companies unlock virtual interactive training for their workforce. Whether it’s training autonomous systems in virtual environments, helping consumers shop with 3D visuals, or scaling tomorrow’s workforce via simulations, I think game tech will infuse every sector in 2025

✈️ Would you trust engineers working with 1980s-era resolution to design a next-gen aircraft? In simulation, mesh cells are like pixels: the more you have, the more detail you capture. But until now, even advanced engineering teams were limited to coarse meshes—far from the fidelity needed to fully trust or iterate on complex systems. 🛠️ Simulations at industrial scale have long been the bottleneck in designing and operating complex systems—from aircraft to cars to energy infrastructure. For decades, most of the improvement has been coming from Moore's law - CPU speed improvements, rather than software (some of which still use Fortran, an insider admitted during one of my many reference calls...) Enters AI and an amazing team who've been obsessed with that problem for year. So far, most models have broken down beyond meshes with a few 100K cells on multiple GPUs. But yesterday, the Emmi AI team released AB-UPT, the first fluid dynamics model scaling beyond 150 million mesh cells, running on a single GPU, and delivering real-world physics accuracy. - 150 million mesh cells (no typo, that's 1000x more) - Real-world accuracy on a single GPU This is not just faster simulation—it’s AI-native simulation that finally bridges the gap between research demos and industrial-grade engineering. It means aircraft designers can explore concepts in minutes instead of weeks. It means complex systems can be simulated in real time as they operate. At Serena Data Ventures, together with Juliette, Matthieu, Floriane, Bertrand and Charline, we invest in foundational technologies that shift what’s possible in infrastructure and foundational software. Johannes, Dennis, Miks and their whole team are doing just that—and doing it fast. Hats off to all of them for this game-changing breakthrough! 📄 𝗥𝗲𝗮𝗱 𝘁𝗵𝗲 𝗽𝗮𝗽𝗲𝗿: https://lnkd.in/deX_zQWx 🤗 𝗧𝗿𝘆 𝘁𝗵𝗲 𝗺𝗼𝗱𝗲𝗹: https://lnkd.in/dmd-xtpR 💻 𝗔𝗰𝗰𝗲𝘀𝘀 𝘁𝗵𝗲 𝗰𝗼𝗱𝗲: https://lnkd.in/dZY6EW_P 🧪 𝗖𝗵𝗲𝗰𝗸 𝘁𝗵𝗲 𝗱𝗲𝗺𝗼: https://demo.emmi.ai/ #FoundationalAI #CFD

🚀 New preprint alert! Proud to share our latest work: "Thermodynamically Consistent Latent Dynamics Identification for Parametric Systems" 📄 https://lnkd.in/gPCQHYiZ In this paper, we propose tLaSDI, a novel framework for reduced-order modeling that fuses thermodynamic principles with machine learning to model complex, parametric dynamical systems with #interpretability, #consistency, and #speed. 🔍 Key innovations: + #pGFINNs: A new class of GENERIC-informed neural networks that enforce the first and second laws of thermodynamics in latent space dynamics. + #Physics-#informed #active #learning: An adaptive sampling strategy that drastically improves accuracy and efficiency using a physics-informed error indicator. + #Massive #computational #gains: Up to 3,528× speed-up with only 1–3% error, plus 50–90% training cost reduction over prior state-of-the-art. + #Insightful #latent #dynamics: Latent variables reflect #free #energy #conservation and #entropy #generation, offering physically meaningful interpretation of learned models. 🧪 Benchmarks, demonstrating both predictive accuracy and thermodynamic fidelity, include: + Burgers’ equation + 1D/1V Vlasov–Poisson equation 🤝 With amazing collaborators: Xiaolong He, Yeonjong Shin, Anthony Gruber, Sohyeon Jung & Kookjin Lee #neural #network #ML #AI #simulation

Found an exciting new study on 3D modeling, AI and robotics. I'll explain the tech, but first... a story: Imagine pointing a camera at your factory floor or a complex assembly line. Instantly, on your screen, you see a live, interactive 3D model of that entire space – not just the machinery, but also your workers moving within it, all updated continuously in real-time. Think of it like having a perfect, living dynamic dollhouse version of your operations that mirrors reality second-by-second. Rather than a recording of something that already happened, it's live spatial understanding. That's what this new research potentially makes possible. It introduces a framework for simultaneously tracking camera movement, estimating human poses, and reconstructing both the human and the surrounding scene in 3D, all in real-time. Using 3D Gaussian Splatting, it efficiently models dynamic elements. This sets a precedent for creating live, detailed digital twins of humans interacting with environments, which will be crucial for advancements in robotics (so they have real-time perception), virtual and/or augmented reality, and human-computer interaction. Eventually, this means a lot of positive knock-on effects: - Smarter Robots: Robots could use this live 3D view to navigate complex, changing environments and work much more safely and effectively alongside your human workforce. - Hyper-Realistic Training: You could drop trainees into virtual or AR simulations that perfectly replicate live operational conditions for unparalleled realism. - Remote Expertise: Remote experts could literally "walk through" the live digital twin to troubleshoot issues or guide on-site staff with complete, real-time context. This will enable bridging the gap between the physical world and digital systems instantly, enabling much smarter automation, collaboration, and analysis. Paper: https://lnkd.in/eH6VmmCg

D-Wave researchers have published findings in Science demonstrating that their quantum annealing processors can simulate quantum spin glass dynamics more efficiently than leading classical methods. The study shows their quantum computers can perform simulations in minutes that would reportedly take classical supercomputers millions of years, marking a significant step toward practical quantum advantage in scientific applications. #QuantumComputing #DWave #QuantumSimulation #SpinGlass #ComputationalPhysics #QuantumSupremecy

When I worked on sprays in grad school, I never dreamed of this type of simulation, especially with a commercial CFD code. The advancements in Volume of Fluid (VOF) modeling and Adaptive Mesh Refinement (AMR) are truly remarkable. Take a look at this simulation showcasing a liquid jet. The first view reveals the jet overlaid by the Q-criterion isosurface, illustrating vortices colored by streamwise vorticity (blue for negative and red for positive vorticity). This slowed-down view highlights the formation of Kelvin-Helmholtz instabilities and hairpin vortices. The second view shows how AMR accurately captures the liquid-gas interface while keeping computational costs in check. #cfd #convergecfd

🚦 What if you could turn thousands of real-world driving scenes into billions of virtual ones? That’s exactly what Nvidia is doing with tools like Omniverse, Cosmos, and AI-driven simulations. Here’s how it works: 🌍 OmniMap creates geo-accurate 3D environments using maps and satellite data. 🔍 Neural Reconstruction Engine uses AV sensor logs to generate lifelike 4D simulations of real-world conditions. 🧠 Cosmos generates limitless driving scenarios with simple text prompts, allowing for faster, more scalable training. By blending advanced AI with 3D content creation, this tech is transforming industries, using immersive simulations to tackle real-world challenges like autonomous mobility, complex traffic, and weather scenarios, all within safe virtual environments. 🚘 What role do you see 3D simulations playing in the future of transportation? 👉 Follow me at Tom Zerega for the latest in tech and innovation.

Digital twins are a trending topic in healthcare, but where are we now with this approach? 🧬⚙️ I recently came across a solid overview in the Where Tech Meets Bio newsletter outlining how digital twin technology is being applied, experimentally, in clinical trials, medical imaging, and surgical planning (link in the comments) 👇 First, what are digital twins? Simply put, these models are computational replicas of physiological systems, ranging from single organs to entire patients, built using multi-modal data like genomics, imaging, sensor streams, and clinical records. They're designed to simulate biological function, predict outcomes, and optimize decisions across the care continuum. Currently, digital twin models are being tested in several areas, including these: 🔹 Clinical trials: Companies like UnlearnAI use historical patient data and machine learning to create synthetic control arms—simulating disease progression and treatment response without assigning patients to placebo groups. 🔹 Cardiac intervention planning: Philips generates 3D heart models from ultrasound data, integrating patient-specific anatomy with real-time imaging to support catheter-based procedures. 🔹 Organ-level disease modeling: Quibim builds AI-driven twins of the brain, liver, and prostate to extract quantitative biomarkers from MRI, CT, and PET scans—used in drug development and patient stratification. The digital twin stack typically involves: ⚙️ data integration (omics, imaging, IoT) ⚙️ AI/ML modeling (predictive, causal, generative) ⚙️ simulation engines (mechanistic and statistical) ⚙️ deployment via edge/cloud infrastructure for real-time feedback Despite a lot of research and many companies in this field, these technologies are still in the experimental stage, but are gaining traction in systems where physical testing is constrained or ethically limited. Their potential lies in creating continuously updated, individualized models that can support simulation-based decisions in medicine and therapeutics. A good question, when will we have this technology adopted at scale? Time will tell… 🙏 Image credit: BiopharmaTrend

Simulation has long been used to test and optimize the design of various device components, but design spaces are huge and expert human intuition has always been needed to figure out what part of the design space to explore. New AI-based simulation methods are orders of magnitude faster than traditional methods, but the solutions they produce are typically less precise. However, precision is not needed for predicting qualitative results and the AI models are differentiable. These two properties make it faster and easier for a human to explore design spaces, and more importantly they enable higher-level AI tools to replicate a human's intuitive exploration. Here is a link where you can try a simple demo of forward simulation using a physics-informed AI model and compare it to a traditional a finite difference solver: https://lnkd.in/e2UUEi3q