Unified Quantum State Tomography and Hamiltonian Learning Using Transformer Models: A Language-Translation-Like Approach for Quantum Systems

Abstract

Schrödinger's equation serves as a fundamental component in characterizing quantum systems, wherein both quantum state tomography and Hamiltonian learning are instrumental in comprehending and interpreting quantum systems. While numerous techniques exist for carrying out state tomography and learning Hamiltonians individually, no method has been developed to combine these two aspects. In this study, we introduce a new approach that employs the attention mechanism in transformer models to effectively merge quantum state tomography and Hamiltonian learning. By carefully choosing and preparing the training data, our method integrates both tasks without altering the model's architecture, allowing the model to effectively learn the intricate relationships between quantum states and Hamiltonian. We also demonstrate the effectiveness of our approach across various quantum systems, ranging from simple 2-qubit cases to more involved 2D antiferromagnetic Heisenberg structures. The data collection process is streamlined, as it only necessitates a one-way generation process beginning with state tomography. Furthermore, the scalability and few-shot learning capabilities of our method could potentially minimize the resources required for characterizing and optimizing quantum systems. Our research provides valuable insights into the relationship between Hamiltonian structure and quantum system behavior, fostering opportunities for additional studies on quantum systems and the advancement of quantum computation and associated technologies.

arXiv comment15 pages, 10 figures

References in corpus (36)

• Quantum Computing in the NISQ era and beyond
• Quantum sensing
• Scalable multi-particle entanglement of trapped ions
• Predicting Many Properties of a Quantum System from Very Few Measurements
• Quantum state tomography via compressed sensing
• Experimental entanglement of six photons in graph states
• Reconstructing quantum states with generative models
• Continuous quantum error correction via quantum feedback control
• Experimental Quantum Hamiltonian Learning
• Provably efficient machine learning for quantum many-body problems
• Robust Online Hamiltonian Learning
• Efficient estimation of Pauli observables by derandomization
• Quantum metrology for a general Hamiltonian parameter
• Learning a local Hamiltonian from local measurements
• Integrating Neural Networks with a Quantum Simulator for State Reconstruction
• Determining a local Hamiltonian from a single eigenstate
• Sample-efficient learning of quantum many-body systems
• Classical Shadow Tomography with Locally Scrambled Quantum Dynamics
• Machine learning assisted quantum state estimation
• Hamiltonian Tomography via Quantum Quench
• Learning many-body Hamiltonians with Heisenberg-limited scaling
• Quantum state tomography via reduced density matrices
• Local-measurement-based quantum state tomography via neural networks
• Hamiltonian Learning for Quantum Error Correction
• Transformer Quantum State: A Multi-Purpose Model for Quantum Many-Body Problems
• Scalable and Flexible Classical Shadow Tomography with Tensor Networks
• Optimising shadow tomography with generalised measurements
• Robust and Efficient Hamiltonian Learning
• Entanglement guided search for parent Hamiltonians
• Flexible learning of quantum states with generative query neural networks
• From Ground States to Local Hamiltonians
• Quantum imaginary time evolution steered by reinforcement learning
• Supervised learning in Hamiltonian reconstruction from local measurements on eigenstates
• From eigenstate to Hamiltonian: Prospects for ergodicity and localization
• Determining system Hamiltonian from eigenstate measurements without correlation functions
• When can a local Hamiltonian be recovered from a steady state?

Cited by in corpus (6)

• Variational Monte Carlo with Large Patched Transformers
• Dual-Capability Machine Learning Models for Quantum Hamiltonian Parameter Estimation and Dynamics Prediction
• Hamiltonian Learning using Machine Learning Models Trained with Continuous Measurements
• Learning interactions between Rydberg atoms
• Hamiltonian Learning via Inverse Physics-Informed Neural Networks
• Investigating Pure State Uniqueness in Tomography via Optimization