On the Sample Complexity of Quantum Boltzmann Machine Learning

Abstract

Quantum Boltzmann machines (QBMs) are machine-learning models for both classical and quantum data. We give an operational definition of QBM learning in terms of the difference in expectation values between the model and target, taking into account the polynomial size of the data set. By using the relative entropy as a loss function this problem can be solved without encountering barren plateaus. We prove that a solution can be obtained with stochastic gradient descent using at most a polynomial number of Gibbs states. We also prove that pre-training on a subset of the QBM parameters can only lower the sample complexity bounds. In particular, we give pre-training strategies based on mean-field, Gaussian Fermionic, and geometrically local Hamiltonians. We verify these models and our theoretical findings numerically on a quantum and a classical data set. Our results establish that QBMs are promising machine learning models.

Main text: 11 pages, 3 figures. Supplementary information: 16 pages, 2 figures. We correct a mistake that affected both lemmas 6 and 7. We thank Dhrumil Patel and Mark M. Wilde for identifying this mistake

References in corpus (41)

• On the Opportunities and Risks of Foundation Models
• Barren plateaus in quantum neural network training landscapes
• Predicting Many Properties of a Quantum System from Very Few Measurements
• Parameterized quantum circuits as machine learning models
• Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits
• Challenges and Opportunities in Quantum Machine Learning
• Quantum machine learning: a classical perspective
• An initialization strategy for addressing barren plateaus in parametrized quantum circuits
• Quantum Metropolis Sampling
• Searching for collective behavior in a network of real neurons
• Information-theoretic bounds on quantum advantage in machine learning
• An introduction to integrable techniques for one-dimensional quantum systems
• Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps
• Hamiltonian Learning and Certification Using Quantum Resources
• Sampling from the thermal quantum Gibbs state and evaluating partition functions with a quantum computer
• Tomography and Generative Data Modeling via Quantum Boltzmann Training
• Quantum Variational Autoencoder
• Variational Thermal Quantum Simulation via Thermofield Double States
• Sample-efficient learning of quantum many-body systems
• Quantum Belief Propagation
• Quantum-assisted Helmholtz machines: A quantum-classical deep learning framework for industrial datasets in near-term devices
• Barren plateaus in quantum tensor network optimization
• Bias and Generalization in Deep Generative Models: An Empirical Study
• Better Theory for SGD in the Nonconvex World
• A Variational Quantum Algorithm for Preparing Quantum Gibbs States
• Quantum machine learning and quantum biomimetics: A perspective
• Predicting Gibbs-State Expectation Values with Pure Thermal Shadows
• On the complexity of quantum partition functions
• Generative training of quantum Boltzmann machines with hidden units
• Optimal learning of quantum Hamiltonians from high-temperature Gibbs states
• A performance characterization of quantum generative models
• Quantum-assisted associative adversarial network: Applying quantum annealing in deep learning
• Variational Gibbs State Preparation on NISQ devices
• Learning quantum many-body systems from a few copies
• Quantum Thermal State Preparation
• Handbook of Convergence Theorems for (Stochastic) Gradient Methods
• The Complexity of Making the Gradient Small in Stochastic Convex Optimization
• Power of Quantum Generative Learning
• Training Quantum Boltzmann Machines with the $\beta$-Variational Quantum Eigensolver
• Dissipative Quantum Gibbs Sampling
• Efficient learning of ground & thermal states within phases of matter

Cited by in corpus (12)

• Barren Plateaus in Variational Quantum Computing
• Does provable absence of barren plateaus imply classical simulability?
• Trainability barriers and opportunities in quantum generative modeling
• Evolved Quantum Boltzmann Machines
• Scalable Quantum Computational Science: A Perspective from Block-Encodings and Polynomial Transformations
• Quantum natural gradient with thermal-state initialization
• Constrained free energy minimization for the design of thermal states and stabilizer thermodynamic systems
• Information geometry of bosonic Gaussian thermal states
• Natural gradient and parameter estimation for quantum Boltzmann machines
• Expressive equivalence of classical and quantum restricted Boltzmann machines
• Quantum Chebyshev Probabilistic Models for Fragmentation Functions
• Provable and Verifiable Quantum Advantage in Sample Complexity