Applications of AI in Quantum System Analysis

Headline: China’s Oceanlite Supercomputer Marries AI and Quantum Science—37 Million Cores Simulate Molecular Quantum Chemistry Introduction: In a milestone achievement, Chinese researchers have fused artificial intelligence with traditional supercomputing to simulate complex quantum chemistry at molecular scale—without using a quantum computer. Using the Oceanlite supercomputer powered by 37 million processing cores, the Sunway team has achieved a feat previously deemed impossible on classical machines. Key Insights: 1. Bridging AI and Quantum Physics Quantum chemistry models the probabilistic behavior of particles like electrons within molecules, governed by the wavefunction (Ψ). Such simulations are normally restricted to small molecules due to the exponential complexity of quantum states. To overcome this barrier, the Sunway team used neural-network quantum states (NNQS), allowing machine learning to approximate molecular wavefunctions with quantum-level accuracy. 2. Record-Breaking Simulation Researchers modeled a molecular system containing 120 spin orbitals—the largest AI-driven quantum chemistry simulation ever conducted on a classical supercomputer. The NNQS trained to predict electron energy distributions and refined itself iteratively until it mirrored true molecular quantum behavior. This approach demonstrates that deep learning frameworks can replicate quantum effects at unprecedented scale. 3. Oceanlite’s Engineering Triumph The experiment ran on the Sunway SW26010-Pro CPU, each chip featuring 384 cores optimized for high-performance computing (HPC). Engineers built a hierarchical communication model where management cores coordinated millions of lightweight compute processing elements (CPEs). Achieved 92% strong scaling and 98% weak scaling efficiency, indicating near-perfect hardware-software synchronization—an exceptional accomplishment in exascale computing. 4. Strategic and Scientific Impact Marks a leap forward for China’s AI and quantum research sectors, blending HPC power with neural architectures. The achievement positions China at the frontier of simulating quantum systems without quantum hardware. Why It Matters: This breakthrough redefines the boundary between classical and quantum computing, offering a path to simulate and design complex molecules—essential for materials science, drug discovery, and clean energy research—using today’s infrastructure. It also signals China’s deepening command of exascale computing and its integration with AI, setting a new global benchmark in scientific computing innovation. I share daily insights with 28,000+ followers and 10,000+ professional contacts across defense, tech, and policy. If this topic resonates, I invite you to connect and continue the conversation. Keith King https://lnkd.in/gHPvUttw

🔗✨ Exploring the Future of Quantum Computing with Physics-Informed Neural Networks (PINNs) ✨🔗 Excited to highlight the pioneering work by Stefano Markidis that dives deep into the potential of Quantum Physics-Informed Neural Networks (Quantum PINNs) for solving differential equations on hybrid CPU-QPU systems! 📘 What’s this about? Physics-Informed Neural Networks (PINNs) have proven their versatility in addressing scientific computing challenges. This study extends PINNs into the quantum realm using Continuous Variable (CV) Quantum Computing, offering a new approach to solving Partial Differential Equations (PDEs) with quantum hardware. Key Highlights: ✅ Quantum Meets Physics: The framework combines CV quantum neural networks with classical methods to tackle PDEs like the 1D Poisson equation. ✅ Optimizer Insights: Traditional optimizers like SGD outperformed adaptive methods in this quantum landscape, highlighting the unique challenges of quantum optimization. ✅ Scalability: Explores batch processing and neural network depth for more effective performance on quantum systems. ✅ Programming Ease: Tools like Strawberry Fields and TensorFlow simplify the integration of quantum and classical computations. 💡 Why it matters: This research doesn't just apply PINNs to quantum computing—it highlights the differences between classical and quantum approaches, paving the way for advancements in quantum PINN solvers and their real-world applications in computational physics, electromagnetics, and more. 📖 Dive deeper: Access the full study here: https://lnkd.in/dZm3F3CR Source code available: https://lnkd.in/dAsXxnbN What are your thoughts on combining quantum computing with AI for scientific breakthroughs? Let’s discuss! 🚀 #QuantumComputing #PhysicsInformedNeuralNetworks #ScientificComputing #HybridAI #PDEsolvers #Innovation

𝗤𝗨𝗔𝗡𝗧𝗨𝗠 𝗔𝗡𝗗 𝗔𝗜: 𝗪𝗵𝗲𝗿𝗲 𝘄𝗲 𝘀𝘁𝗮𝗻𝗱 (𝗡𝗼𝘃. 𝟮𝟬𝟮𝟱) Yesterday I attended a Dell Technologies/AMD dinner with AI leaders from several companies, and we had time to reflect on AI adoption and its real impact. Naturally, the question came up: what about the connection between quantum and AI? This was timely because a new primer on quantum machine learning was published recently. It does something rare in this field. Instead of simply listing algorithms, it explains 𝘄𝗵𝘆 many proposed quantum speedups rely on assumptions that do not always hold in practice. The authors take the time to peel back the layers. Many quantum ML results look impressive on paper, but only if you have ideal state preparation, perfect access to data, or a very specific problem structure. When these assumptions are relaxed, the quantum advantage can disappear. The paper goes through this step by step and shows where the real bottlenecks are. It is very clear that encoding classical data into quantum states is often the main cost, and this step alone can cancel out theoretical speedups. The most convincing results appear in a different setting: learning directly from quantum data. Here the situation changes. Quantum states, processes, and many-body dynamics are things classical models cannot easily represent or access. In this regime, quantum learners can use fewer samples, rely on collective measurements, and leverage entanglement to extract more information per experiment. Some tasks even show provable gaps between what quantum and classical learners can do. This part of the paper feels very grounded because these techniques are already used in labs for tomography, Hamiltonian learning, and experimental optimization. So the message is not that quantum ML is unrealistic. It is that near-term progress is much more promising on quantum data than on classical datasets. The paper is unusually direct about this, which is why I found it worth sharing. For comparison, I also looked again at an AI-for-quantum white paper that came out a few months ago. On this side the progress is much more immediate. Machine learning is already improving control, calibration, compilation, sensing, simulation, and even early error-decoding strategies. It is incremental engineering work, but it is happening now and it makes a difference. Together, the two perspectives fit well. Quantum for AI is still research-heavy and depends on strong conditions. AI for quantum is already helping current devices perform better. Knowing this distinction gives a more realistic picture of where the impact is coming from today. #AI #QuantumComputing #QML #QuantumAI #DeepTech

The first time I saw machine learning in action for quantum computing was during my time at the Niels Bohr Institute, University of Copenhagen. Anasua Chatterjee and colleagues were exploring AI-driven methods to automate the tune-up of spin qubits. To be honest, I didn’t give it much attention at the time. Fast forward to today, and AI feels like the secret sauce accelerating almost every aspect of quantum computing. Think about it: quantum computing is all about mastering exponentially complex systems. AI thrives in high-dimensional, data-rich environments. This pairing? It’s like finding the perfect dance partner. Here’s what’s exciting: AI isn’t just helping to debug or optimize—it’s diving deep into the heart of quantum research. It’s designing qubits, discovering novel error correction codes, and making circuit synthesis more efficient than ever. Tasks that once took teams of researchers weeks to figure out are now becoming automated, adaptive, and scalable. One example I really like? AI-enhanced quantum error correction. Researchers are using neural networks and transformers to achieve error rates below what traditional methods can manage—and they’re doing it at a fraction of the computational cost. Another idea that’s caught my attention is quantum feedback control using transformers. This approach could change how we stabilize and steer quantum systems in real time by leveraging AI models to predict and counteract noise. The question now is: how long before we see more of these theoretical breakthroughs transition to real hardware? Natalia Ares, is quantum feedback control with transformers already in the works? This is such an exciting direction for quantum control and AI! 📸 Credits: Yuri Alexeev et al. (2024)

Really happy to see the official publication today of our paper in Nature Machine Intelligence: "Machine Learning for Practical Quantum Error Mitigation" Haoran Liao, Derek S. Wang, Iskandar Sitdikov, Ciro Salcedo, Alireza Seif, Zlatko Minev 🔍 Context: Quantum computers progress to outperform classical supercomputers, but quantum errors remain the primary obstacle. Quantum error mitigation offers a solution but at the high cost of added runtime. 🤔 Key Question: Can classical machine learning help us overcome errors in today's quantum computers by lowering mitigation overheads, in practice, on real hardware, at the 100 qubit+ scale? 🔬 Our Findings: Using both simulations and experiments on state-of-art quantum computers (up to 100 qubits), we find that machine learning for quantum error mitigation (ML-QEM) can: - Significantly reduce overheads. - Maintain or even outperform the accuracy of traditional methods. - Deliver nearly noise-free results for quantum algorithms. We tested multiple machine learning models on various quantum circuits and noise profiles. And, by leveraging ML-QEM, we were able to mimic conventional mitigation results for large quantum circuits, but with much less overhead. 🌟 Conclusion: Our research underscores the potential synergy between classical hashtag#ML and hashtag#AI and quantum computing. We're excited about the prospects and further research! 🙌 Big thanks to the dream team and many folks who contributed! Let’s share and discuss the implications of this exciting work! 🌟👇 📄 Paper: Nature Machine Intelligence https://lnkd.in/dGYzC3fq 🔓 Free access: View the paper here https://lnkd.in/dN222X7D 📚 Preprint on arXiv https://lnkd.in/dGbzjtjA 👩💻 Code Repository: Explore on GitHub https://lnkd.in/dcn-xPtm 🎥 Seminar: Watch hashtag#IBM @Qiskit on YouTube here https://lnkd.in/dEPRcMVK https://lnkd.in/e7JFgc3J

“Before you can use a quantum computer, you first need to be able to turn it on.” Research that I carried out during my PhD at Oxford has brought us closer to that goal. I'm pleased to share that our paper titled “Cross-architecture tuning of silicon and SiGe-based quantum devices using machine learning” has been published in Nature Scientific Reports. We developed CATSAI (pronounced: Cats-eye), an algorithm capable of tuning three different semiconductor quantum devices—silicon finFET, Ge/Si nanowire, and Ge/SiGe heterostructure— to double quantum dots, using a single approach Forming double quantum dots in these devices is a key step towards creating qubits, the essential building blocks of quantum computers. Not long ago, it was thought that each device type would need its own specialized algorithm. CATSAI changes that by tuning different devices and revealing the complex hypersurfaces that separate regions where current flows from those where it’s blocked. In some cases, finding a double quantum dot is like finding a needle in a haystack—sometimes in just 0.002% of the search space. CATSAI does this on the order of minutes —far quicker than what would typically be possible manually. I remember when I first tried to tune a double quantum dot at the start of my PhD - it took me two weeks. That became the last time I tried to do it by hand. CATSAI relies on two key strategies: 1. Training a machine learning model to recognize single quantum dot features. 2. Leveraging reliable data on where these single dots are located in voltage space to narrow down the search for double quantum dots. This work wouldn’t have been possible without the support of our co-authors and collaborators at IST Austria and the University of Basel. Special thanks to Natalia Ares, who supervised my PhD research and provided invaluable guidance and support throughout this project. I’m also grateful for the opportunity she gave me to work with such an amazing team and technology. Interested in learning more? You can read the full paper here: https://lnkd.in/e7Vz8We9 The possibilities ahead are vast, and I’m eager to see where AI software for semiconductor quantum devices takes us next!

📐 AI Is Crossing a Line: From Solving Math Problems to Doing Mathematical Research For years, Large Reasoning Models (LRMs) have looked impressive at competition-style math - Olympiad problems, benchmarks, short-form proofs. This paper asks a much harder question: - Can AI contribute meaningfully to real, open-ended mathematical research? Researchers from Tsinghua University introduce AI Mathematician (AIM) - a framework explicitly designed for frontier-level mathematics. Why math research is fundamentally harder than competitions The authors identify 2 core gaps between benchmarks and real research: - Intrinsic complexity: research problems don’t have clean solution paths. They require long chains of reasoning, exploration, backtracking & synthesis. - Procedural rigor: every step must be justified Most LLMs fail on both. What AIM does differently: AIM introduces two key mechanisms tailored to research-grade work: - Exploration Mechanism: encourages long-horizon reasoning, allowing the model to explore extended proof paths instead of converging early on shallow solutions. - Pessimistic Reasonable Verification: a deliberately skeptical verification process that checks whether each reasoning step is genuinely warranted This directly targets the failure modes that make LLM-generated proofs unreliable. What AIM actually achieved: The system was evaluated on real mathematical research problems: - Quantum algorithm problems - Absorbing boundary condition analysis - High-contrast limit homogenization Across these domains, AIM was able to: - Autonomously construct substantial portions of formal proofs - Derive non-trivial intermediate insights - In some cases, complete the main proof In others, produce partially correct results with instructive reasoning valuable to human researchers This is not “AI giving hints.” It’s AI operating inside the research process. The broader implication This work suggests that LRM-based agent systems may evolve into genuine research collaborators, capable of exploring mathematical structure, testing conjectures & accelerating discovery. It’s an early system. But it’s a meaningful step toward automated scientific reasoning. Why this matters for independent quants: Top hedge funds and research institutions already employ teams of world-class mathematicians, engineers, and quantitative researchers. Competing with that headcount directly isn’t realistic for individuals. But systems like AIM point to a different strategy: - using advanced AI agents as force multipliers. For independent quants, this opens the door to: - compressing team-level research capability into individual workflows - accelerating model development, validation, and hypothesis testing - pushing toward more sophisticated strategies with stronger mathematical grounding and risk discipline The edge isn’t brute force - it’s leverage. And that’s where this line of research gets very interesting. 📄 Paper: https://lnkd.in/e4RZs39D

Quantum whispers in the GPU roar For Wall Street, more AI means more GPUs, more datacenters, more cloud contracts. And OpenAI–NVIDIA $100B deal locks it in. But quieter signals from research point to a second axis of scaling: not just more metal, but smarter math. It’s about quantum. Let me give you some notable examples from the last week research: 1. Compression: QKANs and quantum activation functions Paper: Quantum Variational Activation Functions Empower Kolmogorov-Arnold Networks Offers replacing fixed nonlinearities with single-qubit variational circuits (DARUANs). These tiny activations generate exponentially richer frequency spectra → so we get same power with exponentially fewer parameters. Quantum KANs (QKANs), built on this idea, already outperformed MLPs and KANs with 30% fewer parameters. 2. Exactness: Coset sampling for lattice algorithms Paper: Exact Coset Sampling for Quantum Lattice Algorithms Proposes a subroutine that cancels unknown offsets and produces exact, uniform cosets, making subsequent Fourier sampling provably correct. Injecting mathematically guaranteed steps into probabilistic workflows means precision: fewer wasted tokens, fewer dead-end paths, less variance in cost per query. 3. Hybridization: quantum-classical models in practice Paper: Hybrid Quantum-Classical Model for Image Classification These models dropped small quantum layers into classical CNNs, showing that they can train faster and use fewer parameters than classical versions. ▪️ What does this mean for inference scaling? Scaling won’t only mean bigger clusters for bigger models. It might also be about: - extracting more from each parameter - cutting errors at the source - and blending quantum and classical strengths. Notably, this direction is not lost on the companies like NVIDIA. There are several signs: • NVIDIA's CUDA-Q – an open software platform for hybrid quantum-classical programming. • NVIDIA also launched DGX Quantum, a reference architecture linking quantum control systems directly into AI supercomputers.  • They are opening a dedicated quantum research center with hardware partners. • Jensen Huang is aggressively investing into quantum startups like PsiQuantum (just raised $1B, saying it’s computer will be ready in two years), Quantinuum, and QuEra through NVentures - a major strategic shift in 2025, validating quantum's commercial timeline. ▪️ So what we will see:  GPUs will remain central. But quantum ideas will be slipping into the story of inference scaling. They are still early, but it's the new axis worth paying attention to. What do you think about it?

Scientists have used AI to discover an easier method to form quantum entanglement between subatomic particles, paving the way for simpler quantum technologies. When particles such as photons become entangled, they can share quantum properties — including information — regardless of the distance between them. This phenomenon is important in quantum physics and is one of the features that makes quantum computers so powerful. But the bonds of quantum entanglement have typically proven challenging for scientists to form. This is because it requires the preparation of two separate entangled pairs, then measuring the strength of entanglement — called a Bell-state measurement — on a photon from each of the pairs. These measurements cause the quantum system to collapse and leave the two unmeasured photons entangled, despite them never having directly interacted with one another. This process of "entanglement swapping" could be used for quantum teleportation. In a new study, published Dec. 2, 2024 in the journal Physical Review Letters, scientists used PyTheus, an AI tool that has been specifically created for designing quantum-optic experiments. The authors of the paper initially set out to reproduce established protocols for entanglement swapping in quantum communications. However, the AI tool kept producing a much simpler method to achieve quantum entanglement of photons. "The authors were able to train a neural network on a set of complex data that describes how you set up this kind of experiment in many different conditions, and the network actually learned the physics behind it," Sofia Vallecorsa, a research physicist for the quantum technology initiative at CERN, who was not involved in the new research, told Live Science. The AI tool proposed that entanglement could emerge because the paths of photons were indistinguishable: when there are several possible sources the photons could have come from, and if their origins become indistinguishable from one another, then entanglement can be produced between them when none existed before. Although the scientists were initially skeptical of the results, the tool kept returning the same solution, so they tested the theory. By adjusting the photon sources and ensuring they were indistinguishable, the physicists created conditions where detecting photons at certain paths guaranteed that two others emerged entangled. This breakthrough in quantum physics has simplified the process by which quantum entanglement can be formed. In future, it could have implications for the quantum networks used for secure messaging, making these technologies much more feasible. Whether it is practical to scale the technology into a commercially viable process remains to be seen, however, as environmental noise and device imperfections could cause instability in the quantum system. #AI #Quantum #Entanglement Quantum entanglement enables a range of futuristic tech. (Johan Jarnestad/ The Royal Swedish Academy of Sciences)

Exciting breakthrough at the intersection of AI and quantum physics! 🧠💻⚛️ I've been diving deep into how we can interpret transformer models through the lens of quantum many-body problems, and the results are mind-blowing. Recent work by Shai et al. (https://lnkd.in/dNDwDenT) shows that transformers encode belief state geometry in their residual stream. Building on this, I've found fascinating parallels with computational graphs approaches as Feynman diagrams in QFT (https://lnkd.in/diZ379br). Building the Bridge: Hidden States as Coupling Constants Recent work by Shai et al. [1] suggests that transformers encode belief state geometry within their residual stream. Here's where the quantum connection gets exciting: these hidden states might hold a key parallel to coupling constants in QFT. Coupling the Analogy: Coupling Constants: In quantum mechanics, coupling constants define the strength of interaction between particles. A higher value signifies a stronger influence. Hidden States as "Effective Coupling Constants": In transformers, the values within the hidden state could be seen as a measure of the "strength" of the connections between the current input and the model's belief about the entire future sequence. Stronger hidden state values could indicate a stronger "belief" or connection between the present input and the model's prediction about the future. Weaker hidden state values might suggest a weaker connection or less influence on the prediction from the current input in the context of the broader future sequence. Key insights: 1. Transformer belief states and Feynman diagrams both exhibit fractal structures 2. Hidden states in transformers correlate with coupling constants in QFT 3. Transformer depth ~ energy scale in renormalization group flow 4. Attention mechanisms ~ interaction propagators in many-body systems This framework opens up new possibilities for optimizing transformer architectures and deepening our understanding of how they capture complex language patterns. To my physicist friends: Imagine treating language as a many-body problem, with words as interacting particles! To my ML colleagues: We might be able to leverage QFT techniques to build more efficient language models! I'm still refining these ideas and would love your input. What implications do you see for your field? Any challenges or opportunities I'm missing? Let's push the boundaries of interdisciplinary science together! 🚀 #AI #QuantumPhysics #NLP #MachineLearning #FeynmanDiagrams