Quantum State Synthesis Methods for Researchers

Interesting new study: "EnQode: Fast Amplitude Embedding for Quantum Machine Learning Using Classical Data." The authors introduce a novel framework to address the limitations of traditional amplitude embedding (AE) [GitHub repo included]. Traditional AE methods often involve deep, variable-length circuits, which can lead to high output error due to extensive gate usage and inconsistent error rates across different data samples. This variability in circuit depth and gate composition results in unequal noise exposure, obscuring the true performance of quantum algorithms. To overcome these challenges, the researchers developed EnQode, a fast AE technique based on symbolic representation. Instead of aiming for exact amplitude representation for each sample, EnQode employs a cluster-based approach to achieve approximate AE with high fidelity. Here are some of the key aspects of EnQode: * Clustering: EnQode begins by using the k-means clustering algorithm to group similar data samples. For each cluster, a mean state is calculated to represent the central characteristics of the data distribution within that cluster. * Hardware-optimized ansatz: For each cluster's mean state, a low-depth, machine-optimized ansatz is trained, tailored to the specific quantum hardware being used (e.g., IBM quantum devices). * Transfer Learning for fast embedding: Once the cluster models are trained offline, transfer learning is used for rapid amplitude embedding of new data samples. An incoming sample is assigned to the nearest cluster, and its embedding circuit is initialized with the optimized parameters of that cluster's mean state. These parameters can then be fine-tuned, significantly accelerating the embedding process without retraining from scratch. * Reduced circuit complexity: EnQode achieved an average reduction of over 28× in circuit depth, over 11× in single-qubit gate count, and over 12× in two-qubit gate count, with zero variability across samples due to its fixed ansatz design. * Higher state fidelity in noisy environments: In noisy IBM quantum hardware simulations, EnQode showed a state fidelity improvement of over 14× compared to the baseline, highlighting its robustness to hardware noise. While the baseline achieved 100% fidelity in ideal simulations (as it performs exact embedding), EnQode maintained an average of 89% fidelity when transpiled to real hardware in ideal simulations, which is considered a good approximation given the significant reduction in circuit complexity. Here the article: https://lnkd.in/dQMbNN7b And here the GitHub repo: https://lnkd.in/dbm7q3eJ #qml #datascience #machinelearning #quantum #nisq #quantumcomputing

Reinforcement learning discovers quantum data-encoding circuits for improved quantum machine learning In current quantum computing research, a challenge lies in how best to transform classical data into meaningful quantum inputs for machine learning tasks. Encoding circuits, which map classical features onto quantum states, are key to the power of quantum machine learning. Yet these circuits are often designed with heuristics and guesswork, leading to suboptimal performance. This unmet need has prompted researchers to look for automated ways to generate encoding strategies that are specifically tuned to the problem at hand. Rapp et al. propose a new solution by using a model-based reinforcement learning algorithm to discover data-encoding circuits automatically. Their approach uses a variant of the MuZero method, which constructs quantum circuits layer by layer. Each layer is composed of one- or two-qubit gates that encode data features through linear or nonlinear rotations. At each step, the RL agent picks a gate from a predefined set, guided by a reward based on the circuit’s predictive power in cross-validation. Notably, the model uses an internal “tree search” to limit the number of quantum-circuit evaluations needed, making it feasible for training. Once the best encoding sequences are found, they are inserted into a projected quantum kernel for classification or regression, bypassing the need for large-scale parameter tuning in typical variational approaches. The authors show that these learned circuits consistently outperform both random designs and commonly used reference architectures. In benchmark tasks, including classification and regression, the RL-generated encoding often yielded comparable or slightly better accuracy than classical models, underscoring the potential of specialized quantum embeddings. Perhaps most important, the method identifies multiple well-performing circuits rather than a single fixed layout, revealing a flexibility in design that was previously missing. Such automated, problem-specific circuit generation opens a path for quantum machine learning to be adapted efficiently to varied tasks and hardware constraints. Paper: https://lnkd.in/da3negKT #MachineLearning #ReinforcementLearning #QuantumML #CircuitDesign #Qubit #NeuralNetworks #DataEncoding #AutoML #Innovation #AIforScience #QuantumKernel #NISQ #ModelBasedRL #Benchmark

Yep, we need #ErrorCorrection in #QuantumComputing. And we already have promising approaches for that. But most implementations of them are still done manually. #Software can help automating corresponding constructions. Today, we launched another solution for that at arXiv (joint work with RWTH Aachen University and Forschungszentrum Jülich)! More precisely, we propose a synthesis method for fault-tolerant state preparation circuits that _automatically_ generates state prep circuits _and_ corresponding verification circuits. Both, an exact approach (guaranteeing depth- or gate-optimal circuits) as well as a heuristic are proposed. Numerical evaluations using d=3 and d=5 codes confirm that the generated circuits exhibit the desired scaling of the logical error rates! 👉 Have a look: https://lnkd.in/ewr4h-ai The methods have been implemented in #opensource as part of our Munich Quantum Toolkit. Check that out at https://lnkd.in/eNZeRXiQ. Huge thanks to Tom Peham, Ludwig Schmid, and Lucas Berent from our team for all your efforts and to Markus Müller for the great collaboration! 🙂

Scientists Unlock Simple Way to Create Powerful Quantum States Scientists at the University of Chicago have discovered a simple way to create powerful quantum states that could help power the next generation of quantum sensors and technologies. And the best part? They didn't need complicated new hardware. Instead, the researchers made a small adjustment to a common quantum optics setup called a cavity QED system, where atoms interact with light trapped between mirrors. The twist is that atoms are arranged in pairs, with each pair receiving equal and opposite energy shifts. This reduces the system's symmetry and unlocks a much wider range of highly entangled quantum states. Why does that matter? 🔹 It can generate powerful entangled states using tools already found in many quantum labs 🔹 Researchers can access different quantum states simply by adjusting lasers rather than rebuilding the hardware 🔹 The approach could enable ultra-precise sensing of magnetic fields and gravitational fields 🔹 It combines two things that are usually difficult to achieve together: extreme sensitivity and strong resistance to noise 🔹 It can create exotic states such as the AKLT state, which has long fascinated physicists and may have applications in quantum computing 🔹 The resulting quantum information can be measured using standard Ramsey techniques, avoiding the need for specialized readout methods This work was supported by Q-NEXT, the U.S. Department of Energy (DOE)'s National Quantum Information Science Research Center led by Argonne National Laboratory. The research, published in Physical Review X, is still theoretical for now. However, the team is already discussing experimental demonstrations with other groups and exploring even more ways to arrange atoms and generate new quantum states. Sometimes the biggest breakthroughs don't come from adding more complexity. They come from finding a smarter way to use the tools we already have. #QuantZen #quantum #physics #science