Barren Plateaus in Variational Quantum Computing

Abstract

Variational quantum computing offers a flexible computational paradigm with applications in diverse areas. However, a key obstacle to realizing their potential is the Barren Plateau (BP) phenomenon. When a model exhibits a BP, its parameter optimization landscape becomes exponentially flat and featureless as the problem size increases. Importantly, all the moving pieces of an algorithm -- choices of ansatz, initial state, observable, loss function and hardware noise -- can lead to BPs when ill-suited. Due to the significant impact of BPs on trainability, researchers have dedicated considerable effort to develop theoretical and heuristic methods to understand and mitigate their effects. As a result, the study of BPs has become a thriving area of research, influencing and cross-fertilizing other fields such as quantum optimal control, tensor networks, and learning theory. This article provides a comprehensive review of the current understanding of the BP phenomenon.

24 pages, 10 boxes, updated to published version

References in corpus (27)

• An introduction to quantum machine learning
• Hybrid quantum-classical algorithms and quantum error mitigation
• Quantum Computation of Electronic Transitions using a Variational Quantum Eigensolver
• Variational Quantum Eigensolver for Frustrated Quantum Systems
• Training Saturation in Layerwise Quantum Approximate Optimisation
• MoG-VQE: Multiobjective genetic variational quantum eigensolver
• Variational quantum compiling with double Q-learning
• Quantum-to-Classical Correspondence and Hubbard-Stratonovich Dynamical Systems, a Lie-Algebraic Approach
• On fundamental aspects of quantum extreme learning machines
• Variational quantum simulation: a case study for understanding warm starts
• Alleviating Barren Plateaus in Parameterized Quantum Machine Learning Circuits: Investigating Advanced Parameter Initialization Strategies
• Learning shallow quantum circuits
• Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer
• Shortcuts to Quantum Approximate Optimization Algorithm
• Quantum Convolutional Neural Networks are Effectively Classically Simulable
• Computing exact moments of local random quantum circuits via tensor networks
• Beyond unital noise in variational quantum algorithms: noise-induced barren plateaus and limit sets
• Constrained and Vanishing Expressivity of Quantum Fourier Models
• The role of data embedding in equivariant quantum convolutional neural networks
• Trainability Barriers in Low-Depth QAOA Landscapes
• Noise-induced shallow circuits and absence of barren plateaus
• F-Divergences and Cost Function Locality in Generative Modelling with Quantum Circuits
• Equivalence of cost concentration and gradient vanishing for quantum circuits: An elementary proof in the Riemannian formulation
• Barren plateaus are swamped with traps
• Beyond Quantum Annealing: Optimal control solutions to MaxCut problems
• Efficient quantum-enhanced classical simulation for patches of quantum landscapes
• Dynamic parameterized quantum circuits: expressive and barren-plateau free

Cited by in corpus (89)

• Artificial Intelligence for Quantum Computing
• Classically estimating observables of noiseless quantum circuits
• Quantum Convolutional Neural Networks are Effectively Classically Simulable
• Scalable Quantum Simulations of Scattering in Scalar Field Theory on 120 Qubits
• Neural quantum kernels: training quantum kernels with quantum neural networks
• Constrained and Vanishing Expressivity of Quantum Fourier Models
• Quantum DeepONet: Neural operators accelerated by quantum computing
• Efficient Quantum Circuit Compilation for Near-Term Quantum Advantage
• Adversarial Robustness Guarantees for Quantum Classifiers
• Quantum continual learning on a programmable superconducting processor
• Approximate Unitary $k$-Designs from Shallow, Low-Communication Circuits
• Quantum Kernel Learning for Small Dataset Modeling in Semiconductor Fabrication: Application to Ohmic Contact
• Quantum reservoir computing on random regular graphs
• Double-bracket quantum algorithms for quantum imaginary-time evolution
• Towards Variational Quantum Algorithms for generalized linear and nonlinear transport phenomena
• Analyzing the quantum approximate optimization algorithm: ansätze, symmetries, and Lie algebras
• Quantum computing for genomics: conceptual challenges and practical perspectives
• Quantum Dynamics Simulation of the Advection-Diffusion Equation
• Compressed space quantum approximate optimization algorithm for constrained combinatorial optimization
• Scalability Challenges in Variational Quantum Optimization under Stochastic Noise
• Energy Landscape Plummeting in Variational Quantum Eigensolver: Subspace Optimization, Non-iterative Corrections and Generator-informed Initialization for Improved Quantum Efficiency
• Efficient quantum-enhanced classical simulation for patches of quantum landscapes
• On the similarity of bandwidth-tuned quantum kernels and classical kernels
• Deep Circuit Compression for Quantum Dynamics via Tensor Networks
• Pitfalls when tackling the exponential concentration of parameterized quantum models
• Architectures and random properties of symplectic quantum circuits
• Almost fault-tolerant quantum machine learning with drastic overhead reduction
• Certified Random Number Generation using Quantum Computers
• Optimization via Quantum Preconditioning
• Classical Benchmarks for Variational Quantum Eigensolver Simulations of the Hubbard Model
• Exploring nontrivial topology at quantum criticality in a superconducting processor
• Exact spectral gaps of random one-dimensional quantum circuits
• Harnessing Quantum Computing for Energy Materials: Opportunities and Challenges
• Inequality constraints in variational quantum circuits with qudits
• Non-linear classification capability of quantum neural networks due to emergent quantum metastability
• An Efficient Decomposition of the Carleman Linearized Burgers' Equation
• Bridging Quantum Computing and Nuclear Structure: Atomic Nuclei on a Trapped-Ion Quantum Computer
• Quantum generative modeling for financial time series with temporal correlations
• Toward scalable quantum computations of atomic nuclei
• Exponential distillation of dominant eigenproperties
• Quantum Physics-Informed Neural Networks for Maxwell's Equations: Circuit Design, "Black Hole" Barren Plateaus Mitigation, and GPU Acceleration
• Regularizing quantum loss landscapes by noise injection
• The Lie Algebra of XY-mixer Topologies and Warm Starting QAOA for Constrained Optimization
• Quantitative convergence of trained quantum neural networks to a Gaussian process
• Fighting Exponentially Small Gaps by Counterdiabatic Driving
• Quantum natural gradient with thermal-state initialization
• Generative flow-based warm start of the variational quantum eigensolver
• The Dual Role of Low-Weight Pauli Propagation: A Flawed Simulator but a Powerful Initializer for Variational Quantum Algorithms
• Estimates of loss function concentration in noisy parametrized quantum circuits
• Efficient Estimation and Sequential Optimization of Cost Functions in Variational Quantum Algorithms
• Optimal quantum reservoir learning in proximity to universality
• Graph Coloring via Quantum Optimization on a Rydberg-Qudit Atom Array
• An Error Mitigated Non-Orthogonal Quantum Eigensolver via Shadow Tomography
• Weighted Approximate Quantum Natural Gradient for Variational Quantum Eigensolver
• Approximate Quadratization of High-Order Hamiltonians for Combinatorial Quantum Optimization
• Resource-efficient quantum algorithm for linear systems of equations
• Guided Graph Compression for Quantum Graph Neural Networks
• Exploiting many-body localization for scalable variational quantum simulation
• Double descent in quantum kernel methods
• Notes on Quantum Computing for Thermal Science
• Decoded Quantum Interferometry Under Noise
• A penalty-free quantum algorithm to find energy eigenstates
• Evidence for effectively constant shot complexity in the quantum approximate optimization algorithm without per-instance optimization
• Slice-Wise Initial State Optimization to Improve Cost and Accuracy of the VQE on Lattice Models
• Quantum approximate optimization of finite-state bosonic systems
• Recent quantum runtime (dis)advantages
• Double-bracket quantum algorithms for high-fidelity ground state preparation
• Quantum feature-map learning with reduced resource overhead
• Adaptive time Compressed QITE (ACQ) and its geometrical interpretation
• Qubit-efficient quantum combinatorial optimization solver
• Performance evaluation of variational quantum eigensolver and quantum dynamics algorithms on the advection-diffusion equation
• Moments of Quantum Channel Ensembles
• Exploring the performance of superposition of product states: from 1D to 3D quantum spin systems
• Interpolation-based coordinate descent method for parameterized quantum circuits
• Exploring fixed points and eigenstates of quantum systems with reinforcement learning
• Quantum Encoding of Structured Data with Matrix Product States
• Designing a Machine Learning-Driven, Cross-Hardware Emulator for Noisy Quantum Computers with Gate-Based Protocols
• Quantum Simulation-Based Optimization for Cooling System Design
• Equivalence between exponential concentration in quantum machine learning kernels and barren plateaus in variational algorithms
• Direct Gradient Computation for Barren Plateaus in Parameterized Quantum Circuits
• Who can compete with quantum computers? Lecture notes on quantum inspired tensor networks computational techniques
• Learning complexity gradually in quantum machine learning models
• Gradient Scalability and Taylor Surrogation of Quantum Cost Landscapes
• Variational Quantum Subspace Construction via Symmetry-Preserving Cost Functions
• Resource-efficient Quantum Algorithms for Selected Hamiltonian Subspace Diagonalization
• Optimizing Unitary Coupled Cluster Wave Functions on Quantum Hardware: Error Bound and Resource-Efficient Optimizer
• Role of overparametrization in quantum approximate optimization
• Preparation of the single-spinon wave function on a quantum computer
• Superior resilience to poisoning and amenability to unlearning in quantum machine learning