On barren plateaus and cost function locality in variational quantum algorithms

Abstract

Variational quantum algorithms rely on gradient based optimization to iteratively minimize a cost function evaluated by measuring output(s) of a quantum processor. A barren plateau is the phenomenon of exponentially vanishing gradients in sufficiently expressive parametrized quantum circuits. It has been established that the onset of a barren plateau regime depends on the cost function, although the particular behavior has been demonstrated only for certain classes of cost functions. Here we derive a lower bound on the variance of the gradient, which depends mainly on the width of the circuit causal cone of each term in the Pauli decomposition of the cost function. Our result further clarifies the conditions under which barren plateaus can occur.

arXiv comment26 pages, RevTeX

References in corpus (25)

• A variational eigenvalue solver on a quantum processor
• Hardware-efficient Variational Quantum Eigensolver for Small Molecules and Quantum Magnets
• Supervised learning with quantum enhanced feature spaces
• Barren plateaus in quantum neural network training landscapes
• The theory of variational hybrid quantum-classical algorithms
• Quantum Circuit Learning
• Evaluating analytic gradients on quantum hardware
• An adaptive variational algorithm for exact molecular simulations on a quantum computer
• Circuit-centric quantum classifiers
• Time Optimal Control in Spin Systems
• Towards Practical Quantum Variational Algorithms
• Integration with respect to the Haar measure on unitary, orthogonal and symplectic group
• Self-Verifying Variational Quantum Simulation of the Lattice Schwinger Model
• Measurement-driven entanglement transition in hybrid quantum circuits
• Quantum autoencoders for efficient compression of quantum data
• An initialization strategy for addressing barren plateaus in parametrized quantum circuits
• Quantum-assisted quantum compiling
• Quantum algorithms for electronic structure calculations: particle/hole Hamiltonian and optimized wavefunction expansions
• Local random quantum circuits are approximate polynomial-designs
• Towards Quantum Machine Learning with Tensor Networks
• Measurement Optimization in the Variational Quantum Eigensolver Using a Minimum Clique Cover
• Reachability Deficits in Quantum Approximate Optimization
• Convergence Conditions for Random Quantum Circuits
• Scaling of variational quantum circuit depth for condensed matter systems
• Variationally Learning Grover's Quantum Search Algorithm

Cited by in corpus (75)

• Variational Quantum Algorithms
• The Variational Quantum Eigensolver: a review of methods and best practices
• Noise-Induced Barren Plateaus in Variational Quantum Algorithms
• Challenges and Opportunities in Quantum Machine Learning
• Generalization in quantum machine learning from few training data
• Effect of barren plateaus on gradient-free optimization
• Diagnosing Barren Plateaus with Tools from Quantum Optimal Control
• Barren Plateaus in Variational Quantum Computing
• Training Quantum Embedding Kernels on Near-Term Quantum Computers
• Equivalence of quantum barren plateaus to cost concentration and narrow gorges
• A Lie Algebraic Theory of Barren Plateaus for Deep Parameterized Quantum Circuits
• Entanglement Devised Barren Plateau Mitigation
• Theory for Equivariant Quantum Neural Networks
• Recent advances for quantum classifiers
• A comprehensive review of Quantum Machine Learning: from NISQ to Fault Tolerance
• Theoretical Guarantees for Permutation-Equivariant Quantum Neural Networks
• Does provable absence of barren plateaus imply classical simulability?
• Exact and efficient Lanczos method on a quantum computer
• On the practical usefulness of the Hardware Efficient Ansatz
• Simulating Hadronic Physics on NISQ devices using Basis Light-Front Quantization
• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Absence of barren plateaus in finite local-depth circuits with long-range entanglement
• Variational Quantum Classifiers Through the Lens of the Hessian
• Dynamical simulation via quantum machine learning with provable generalization
• Two quantum algorithms for solving the one-dimensional advection-diffusion equation
• Quantum Mixed State Compiling
• Quantum Capsule Networks
• On fundamental aspects of quantum extreme learning machines
• Contextual Subspace Variational Quantum Eigensolver
• Engineered dissipation to mitigate barren plateaus
• Tight and Efficient Gradient Bounds for Parameterized Quantum Circuits
• Reduction of finite sampling noise in quantum neural networks
• Training variational quantum algorithms with random gate activation
• Efficient estimation of trainability for variational quantum circuits
• Calculating nonadiabatic couplings and Berry's phase by variational quantum eigensolvers
• Quantum Monte Carlo for Economics: Stress Testing and Macroeconomic Deep Learning
• Quantum Error Mitigation Relying on Permutation Filtering
• The quantum cost function concentration dependency on the parametrization expressivity
• Variational quantum solver employing the PDS energy functional
• Iterative Layerwise Training for Quantum Approximate Optimization Algorithm
• Reducing the Resources Required by ADAPT-VQE Using Coupled Exchange Operators and Improved Subroutines
• Mitigated barren plateaus in the time-nonlocal optimization of analog quantum-algorithm protocols
• Dissipative variational quantum algorithms for Gibbs state preparation
• Experimental Quantum Learning of a Spectral Decomposition
• Robustness of Variational Quantum Algorithms against stochastic parameter perturbation
• Coarse grained intermolecular interactions on quantum processors
• Fermionic Simulators for Enhanced Scalability of Variational Quantum Simulation
• F-Divergences and Cost Function Locality in Generative Modelling with Quantum Circuits
• Resource frugal optimizer for quantum machine learning
• Characterization of variational quantum algorithms using free fermions
• Barren plateaus are swamped with traps
• Absence of barren plateaus and scaling of gradients in the energy optimization of isometric tensor network states
• Applicability of Measurement-based Quantum Computation towards Physically-driven Variational Quantum Eigensolver
• Digital-analog quantum learning on Rydberg atom arrays
• Variational quantum computing for quantum simulation: principles, implementations, and challenges
• Simulating thermodynamic properties of dinuclear metal complexes using Variational Quantum Algorithms
• Optimal-order Trotter-Suzuki decomposition for quantum simulation on noisy quantum computers
• Solving lattice gauge theories using the quantum Krylov algorithm and qubitization
• Reusability Report: Comparing gradient descent and monte carlo tree search optimization of quantum annealing schedules
• Machine-Learning Insights into the Entanglement-trainability Correlation of Parameterized Quantum Circuits
• Quantum Inception Score
• Quantum circuits for partial differential equations in Fourier space
• Universal compiling and (No-)Free-Lunch theorems for continuous variable quantum learning
• Perturbative gadgets for gate-based quantum computing: Non-recursive constructions without subspace restrictions
• Almost fault-tolerant quantum machine learning with drastic overhead reduction
• Encoding strongly-correlated many-boson wavefunctions on a photonic quantum computer: application to the attractive Bose-Hubbard model
• The Dual Role of Low-Weight Pauli Propagation: A Flawed Simulator but a Powerful Initializer for Variational Quantum Algorithms
• Batched Line Search Strategy for Navigating through Barren Plateaus in Quantum Circuit Training
• Estimates of loss function concentration in noisy parametrized quantum circuits
• Direct Gradient Computation for Barren Plateaus in Parameterized Quantum Circuits
• Trainability issues in quantum policy gradients
• Equivalence between exponential concentration in quantum machine learning kernels and barren plateaus in variational algorithms
• Analog Quantum Approximate Optimization Algorithm
• Data-driven Quantum Dynamical Embedding Method for Long-term Prediction on Near-term Quantum Computers
• Noisy Monitored Quantum Circuits