Trainability Enhancement of Parameterized Quantum Circuits via Reduced-Domain Parameter Initialization

Abstract

Parameterized quantum circuits (PQCs) have been widely used as a machine learning model to explore the potential of achieving quantum advantages for various tasks. However, training PQCs is notoriously challenging owing to the phenomenon of plateaus and/or the existence of (exponentially) many spurious local minima. To enhance trainability, in this work we propose an efficient parameter initialization strategy with theoretical guarantees. We prove that by reducing the initial domain of each parameter inversely proportional to the square root of circuit depth, the magnitude of the cost gradient decays at most polynomially with respect to qubit count and circuit depth. Our theoretical results are substantiated through numerical simulations of variational quantum eigensolver tasks. Moreover, we demonstrate that the reduced-domain initialization strategy can protect specific quantum neural networks from exponentially many spurious local minima. Our results highlight the significance of an appropriate parameter initialization strategy, offering insights to enhance the trainability and convergence of variational quantum algorithms.

7 figures; Numerical simulations added

References in corpus (48)

• Quantum Machine Learning
• Variational Quantum Algorithms
• Ising formulations of many NP problems
• Barren plateaus in quantum neural network training landscapes
• The theory of variational hybrid quantum-classical algorithms
• Quantum computational chemistry
• Quantum Circuit Learning
• Quantum Chemistry in the Age of Quantum Computing
• Quantum Convolutional Neural Networks
• Parameterized quantum circuits as machine learning models
• Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits
• From the Quantum Approximate Optimization Algorithm to a Quantum Alternating Operator Ansatz
• Quantum Approximate Optimization Algorithm: Performance, Mechanism, and Implementation on Near-Term Devices
• Connecting ansatz expressibility to gradient magnitudes and barren plateaus
• Classical simulation of quantum many-body systems with a tree tensor network
• Quantum Simulators: Architectures and Opportunities
• A rigorous and robust quantum speed-up in supervised machine learning
• Theory of variational quantum simulation
• Training variational quantum algorithms is NP-hard
• An initialization strategy for addressing barren plateaus in parametrized quantum circuits
• Layerwise learning for quantum neural networks
• Hierarchical quantum classifiers
• Exploring entanglement and optimization within the Hamiltonian Variational Ansatz
• Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps
• Stochastic gradient descent for hybrid quantum-classical optimization
• Quantum machine learning beyond kernel methods
• Trainability of Dissipative Perceptron-Based Quantum Neural Networks
• Theory of overparametrization in quantum neural networks
• Quantum algorithms to simulate many-body physics of correlated fermions
• Avoiding local minima in variational quantum eigensolvers with the natural gradient optimizer
• Large gradients via correlation in random parameterized quantum circuits
• Entanglement Devised Barren Plateau Mitigation
• Efficient variational simulation of non-trivial quantum states
• TensorCircuit: a Quantum Software Framework for the NISQ Era
• Applying the Quantum Approximate Optimization Algorithm to the Tail Assignment Problem
• Analyzing the barren plateau phenomenon in training quantum neural networks with the ZX-calculus
• Analytic theory for the dynamics of wide quantum neural networks
• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Training Saturation in Layerwise Quantum Approximate Optimisation
• Symmetry enhanced variational quantum spin eigensolver
• Improved machine learning algorithm for predicting ground state properties
• Variational circuit compiler for quantum error correction
• End-to-End Quantum Machine Learning Implemented with Controlled Quantum Dynamics
• Variational waveguide QED simulators
• Quantum variational learning for quantum error-correcting codes
• Quantum Geometric Machine Learning for Quantum Circuits and Control
• Practical Verification of Quantum Properties in Quantum Approximate Optimization Runs
• EHA: Entanglement-variational Hardware-efficient Ansatz for Eigensolvers

Cited by in corpus (22)

• Barren Plateaus in Variational Quantum Computing
• Does provable absence of barren plateaus imply classical simulability?
• Variational quantum simulation: a case study for understanding warm starts
• Quantum Convolutional Neural Networks are Effectively Classically Simulable
• Barren plateaus are swamped with traps
• Variational quantum computing for quantum simulation: principles, implementations, and challenges
• Adversarial Robustness Guarantees for Quantum Classifiers
• Fast gradient-free optimization of excitations in variational quantum eigensolvers
• Scalability Challenges in Variational Quantum Optimization under Stochastic Noise
• Efficient quantum-enhanced classical simulation for patches of quantum landscapes
• Energy Landscape Plummeting in Variational Quantum Eigensolver: Subspace Optimization, Non-iterative Corrections and Generator-informed Initialization for Improved Quantum Efficiency
• Pitfalls when tackling the exponential concentration of parameterized quantum models
• Expressivity of deterministic quantum computation with one qubit
• Typical Machine Learning Datasets as Low-Depth Quantum Circuits
• Estimates of loss function concentration in noisy parametrized quantum circuits
• Efficient Online Quantum Circuit Learning with No Upfront Training
• A coherent approach to quantum-classical optimization
• Resource-efficient quantum algorithm for linear systems of equations
• Exploiting many-body localization for scalable variational quantum simulation
• Direct Gradient Computation for Barren Plateaus in Parameterized Quantum Circuits
• Learning complexity gradually in quantum machine learning models
• Gradient Scalability and Taylor Surrogation of Quantum Cost Landscapes