Quantum Variational Autoencoder

Abstract

Variational autoencoders (VAEs) are powerful generative models with the salient ability to perform inference. Here, we introduce a quantum variational autoencoder (QVAE): a VAE whose latent generative process is implemented as a quantum Boltzmann machine (QBM). We show that our model can be trained end-to-end by maximizing a well-defined loss-function: a 'quantum' lower-bound to a variational approximation of the log-likelihood. We use quantum Monte Carlo (QMC) simulations to train and evaluate the performance of QVAEs. To achieve the best performance, we first create a VAE platform with discrete latent space generated by a restricted Boltzmann machine (RBM). Our model achieves state-of-the-art performance on the MNIST dataset when compared against similar approaches that only involve discrete variables in the generative process. We consider QVAEs with a smaller number of latent units to be able to perform QMC simulations, which are computationally expensive. We show that QVAEs can be trained effectively in regimes where quantum effects are relevant despite training via the quantum bound. Our findings open the way to the use of quantum computers to train QVAEs to achieve competitive performance for generative models. Placing a QBM in the latent space of a VAE leverages the full potential of current and next-generation quantum computers as sampling devices.

v2: published version. 13 pages, 3 figures, 2 tables

References in corpus (29)

• Quantum Machine Learning
• A variational eigenvalue solver on a quantum processor
• Quantum algorithm for solving linear systems of equations
• Surface codes: Towards practical large-scale quantum computation
• Hardware-efficient Variational Quantum Eigensolver for Small Molecules and Quantum Magnets
• Quantum Annealing in the Transverse Ising Model
• A Quantum Adiabatic Evolution Algorithm Applied to Random Instances of an NP-Complete Problem
• Quantum principal component analysis
• An introduction to quantum machine learning
• Theory of Quantum Annealing of an Ising Spin Glass
• Quantum algorithm for systems of linear equations with exponentially improved dependence on precision
• Quantum Data Fitting
• Quantum autoencoders for efficient compression of quantum data
• A blueprint for demonstrating quantum supremacy with superconducting qubits
• Quantum Boltzmann Machine
• Quantum generalisation of feedforward neural networks
• Opportunities and challenges for quantum-assisted machine learning in near-term quantum computers
• Estimation of effective temperatures in quantum annealers for sampling applications: A case study with possible applications in deep learning
• Searching for quantum speedup in quasistatic quantum annealers
• Tunable Quantum Tunneling of Magnetic Domain Walls
• Quantum adiabatic machine learning
• Population Annealing with Weighted Averages: A Monte Carlo Method for Rough Free Energy Landscapes
• Adiabaticity in open quantum systems
• Quantum-assisted Helmholtz machines: A quantum-classical deep learning framework for industrial datasets in near-term devices
• Designing High-Fidelity Single-Shot Three-Qubit Gates: A Machine Learning Approach
• Temperature scaling law for quantum annealing optimizers
• Genetic Algorithms for Digital Quantum Simulations
• Global warming: Temperature estimation in annealers
• Scalable effective temperature reduction for quantum annealers via nested quantum annealing correction

Cited by in corpus (52)

• Perspectives of quantum annealing: Methods and implementations
• Physics-Inspired Optimization for Quadratic Unconstrained Problems Using a Digital Annealer
• Differentiable Learning of Quantum Circuit Born Machine
• The prospects of quantum computing in computational molecular biology
• Variational Quantum State Diagonalization
• Sequential minimal optimization for quantum-classical hybrid algorithms
• Quantum generative adversarial network for generating discrete distribution
• Quantum Machine Learning for Chemistry and Physics
• Quantum Computing at the Frontiers of Biological Sciences
• Noisy intermediate-scale quantum computers
• Beyond-classical computation in quantum simulation
• Quantum Machine Learning: from physics to software engineering
• Biology and medicine in the landscape of quantum advantages
• Item Listing Optimization for E-commerce Websites based on Diversity
• Generative quantum machine learning via denoising diffusion probabilistic models
• Hybrid quantum-classical machine learning for generative chemistry and drug design
• Quantum machine learning and quantum biomimetics: A perspective
• Variational inference with a quantum computer
• Mean field analysis of reverse annealing for code-division multiple-access multiuser detection
• On the Sample Complexity of Quantum Boltzmann Machine Learning
• Quantum-assisted associative adversarial network: Applying quantum annealing in deep learning
• Searching for anomalous quartic gauge couplings at muon colliders using principle component analysis
• Quantum advantage for differential equation analysis
• High-Dimensional Similarity Search with Quantum-Assisted Variational Autoencoder
• Perils of Embedding for Sampling Problems
• Efficient Algorithm for Binary Quadratic Problem by Column Generation and Quantum Annealing
• Embedding Learning in Hybrid Quantum-Classical Neural Networks
• Near-Term Quantum-Classical Associative Adversarial Networks
• Error measurements for a quantum annealer using the one-dimensional Ising model with twisted boundaries
• Dynamics of Order Parameters of Non-stoquastic Hamiltonians in the Adaptive Quantum Monte Carlo Method
• Physical Realization of Measurement Based Quantum Computation
• Quantum subspace alignment for domain adaptation
• Solving systems of Boolean multivariate equations with quantum annealing
• Perils of Embedding for Quantum Sampling
• Searching for gluon quartic gauge couplings at muon colliders using the auto-encoder
• Message-passing algorithm of quantum annealing with nonstoquastic Hamiltonian
• Hybrid Algorithm of Linear Programming Relaxation and Quantum Annealing
• Quantum kernel learning Model constructed with small data
• Individual subject evaluated difficulty of adjustable mazes generated using quantum annealing
• Online calibration scheme for training restricted Boltzmann machines with quantum annealing
• Information loss and run time from practical application of quantum data compression
• Precise Image Generation on Current Noisy Quantum Computing Devices
• Universal noise-precision relations in variational quantum algorithms
• Error-mitigated photonic quantum circuit Born machine
• Quantum annealing with pairs of ${}^2\Sigma$ molecules as qubits
• Hybrid Quantum Neural Networks with Variational Quantum Regressor for Enhancing QSPR Modeling of CO2-Capturing Amine
• Quantum Inception Score
• Tangible reduction in learning sample complexity with large classical samples and small quantum system
• Analysis of the Hopfield Model with Discrete Coupling
• Quantum neural network autoencoder and classifier applied to an industrial case study
• Autoencoder-based analytic continuation method for strongly correlated quantum systems
• Disentangling quantum neural networks for unified estimation of quantum entropies and distance measures