Connecting ansatz expressibility to gradient magnitudes and barren plateaus

Abstract

Parameterized quantum circuits serve as ansätze for solving variational problems and provide a flexible paradigm for programming near-term quantum computers. Ideally, such ansätze should be highly expressive so that a close approximation of the desired solution can be accessed. On the other hand, the ansatz must also have sufficiently large gradients to allow for training. Here, we derive a fundamental relationship between these two essential properties: expressibility and trainability. This is done by extending the well established barren plateau phenomenon, which holds for ansätze that form exact 2-designs, to arbitrary ansätze. Specifically, we calculate the variance in the cost gradient in terms of the expressibility of the ansatz, as measured by its distance from being a 2-design. Our resulting bounds indicate that highly expressive ansätze exhibit flatter cost landscapes and therefore will be harder to train. Furthermore, we provide numerics illustrating the effect of expressiblity on gradient scalings, and we discuss the implications for designing strategies to avoid barren plateaus.

Main text: 10 pages, 4 figures. Appendices: 10 pages, 2 figures

References in corpus (20)

• Supplementary information for "Quantum supremacy using a programmable superconducting processor"
• A Quantum Approximate Optimization Algorithm
• An introduction to quantum machine learning
• The power of quantum neural networks
• Noise-Induced Barren Plateaus in Variational Quantum Algorithms
• Evenly distributed unitaries: on the structure of unitary designs
• Absence of Barren Plateaus in Quantum Convolutional Neural Networks
• Layerwise learning for quantum neural networks
• Large gradients via correlation in random parameterized quantum circuits
• Variational Quantum State Eigensolver
• Barren plateaus preclude learning scramblers
• Quantum Assisted Simulator
• Learning to learn with quantum neural networks via classical neural networks
• Reformulation of the No-Free-Lunch Theorem for Entangled Data Sets
• Unitary designs from statistical mechanics in random quantum circuits
• Iterative Quantum Assisted Eigensolver
• Variational Hamiltonian Diagonalization for Dynamical Quantum Simulation
• Abrupt Transitions in Variational Quantum Circuit Training
• Toward Trainability of Quantum Neural Networks
• No Free Lunch for Quantum Machine Learning

Cited by in corpus (312)

• Variational Quantum Algorithms
• The Variational Quantum Eigensolver: a review of methods and best practices
• The power of quantum neural networks
• Noise-Induced Barren Plateaus in Variational Quantum Algorithms
• Challenges and Opportunities in Quantum Machine Learning
• Generalization in quantum machine learning from few training data
• Absence of Barren Plateaus in Quantum Convolutional Neural Networks
• Effect of barren plateaus on gradient-free optimization
• Diagnosing Barren Plateaus with Tools from Quantum Optimal Control
• Barren Plateaus in Variational Quantum Computing
• Quantum Computing for High-Energy Physics: State of the Art and Challenges. Summary of the QC4HEP Working Group
• Quantum circuit architecture search for variational quantum algorithms
• Exploiting symmetry in variational quantum machine learning
• Supervised quantum machine learning models are kernel methods
• Theory of overparametrization in quantum neural networks
• Group-Invariant Quantum Machine Learning
• Quantum annealing initialization of the quantum approximate optimization algorithm
• Equivalence of quantum barren plateaus to cost concentration and narrow gorges
• A Lie Algebraic Theory of Barren Plateaus for Deep Parameterized Quantum Circuits
• Introduction to Haar Measure Tools in Quantum Information: A Beginner's Tutorial
• Hybrid quantum-classical algorithms in the noisy intermediate-scale quantum era and beyond
• Higher Order Derivatives of Quantum Neural Networks with Barren Plateaus
• Efficient measure for the expressivity of variational quantum algorithms
• Quantum computing models for artificial neural networks
• Scalable measures of magic resource for quantum computers
• Hybrid Quantum-Classical Generative Adversarial Network for High Resolution Image Generation
• Quantum Machine Learning: from physics to software engineering
• 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
• Layer VQE: A Variational Approach for Combinatorial Optimization on Noisy Quantum Computers
• Does provable absence of barren plateaus imply classical simulability?
• Analyzing the barren plateau phenomenon in training quantum neural networks with the ZX-calculus
• Understanding quantum machine learning also requires rethinking generalization
• Towards a Larger Molecular Simulation on the Quantum Computer: Up to 28 Qubits Systems Accelerated by Point Group Symmetry
• On the practical usefulness of the Hardware Efficient Ansatz
• Digitized-Counterdiabatic Quantum Algorithm for Protein Folding
• Hamiltonian variational ansatz without barren plateaus
• Towards adiabatic quantum computing using compressed quantum circuits
• Subtleties in the trainability of quantum machine learning models
• Avoiding barren plateaus via transferability of smooth solutions in Hamiltonian Variational Ansatz
• Constrained Quantum Optimization for Extractive Summarization on a Trapped-ion Quantum Computer
• Synergy Between Quantum Circuits and Tensor Networks: Short-cutting the Race to Practical Quantum Advantage
• Machine Learning for Anomaly Detection in Particle Physics
• Analytic theory for the dynamics of wide quantum neural networks
• Generalization despite overfitting in quantum machine learning models
• Building spatial symmetries into parameterized quantum circuits for faster training
• Quantum variational optimization: The role of entanglement and problem hardness
• Variational inference with a quantum computer
• Reflection Equivariant Quantum Neural Networks for Enhanced Image Classification
• Avoiding Barren Plateaus with Classical Deep Neural Networks
• Can Error Mitigation Improve Trainability of Noisy Variational Quantum Algorithms?
• Challenges of variational quantum optimization with measurement shot noise
• Gradient Ascent Pulse Engineering with Feedback
• Trainability barriers and opportunities in quantum generative modeling
• Mitigating Barren Plateaus with Transfer-learning-inspired Parameter Initializations
• Ancilla-free implementation of generalized measurements for qubits embedded in a qudit space
• NP-hard but no longer hard to solve? Using quantum computing to tackle optimization problems
• Absence of barren plateaus in finite local-depth circuits with long-range entanglement
• Quantum Phase Recognition via Quantum Kernel Methods
• Provably Trainable Rotationally Equivariant Quantum Machine Learning
• Quantum Error Correction with Quantum Autoencoders
• Long-time simulations with high fidelity on quantum hardware
• Trainability Enhancement of Parameterized Quantum Circuits via Reduced-Domain Parameter Initialization
• An Expressive Ansatz for Low-Depth Quantum Approximate Optimisation
• Dynamical simulation via quantum machine learning with provable generalization
• Interpretable Quantum Advantage in Neural Sequence Learning
• Variational Quantum Approximate Support Vector Machine with Inference Transfer
• A semi-agnostic ansatz with variable structure for quantum machine learning
• Exponential data encoding for quantum supervised learning
• Fock State-enhanced Expressivity of Quantum Machine Learning Models
• Nonlinear dynamics as a ground-state solution on quantum computers
• Quantum Mixed State Compiling
• Challenges in the use of quantum computing hardware-efficient Ansatze in electronic structure theory
• Entangled Datasets for Quantum Machine Learning
• Entanglement entropy production in Quantum Neural Networks
• Variational waveguide QED simulators
• On fundamental aspects of quantum extreme learning machines
• Quantum algorithms from fluctuation theorems: Thermal-state preparation
• Classification of dynamical Lie algebras for translation-invariant 2-local spin systems in one dimension
• Variational quantum simulation: a case study for understanding warm starts
• An Empirical Comparison of Optimizers for Quantum Machine Learning with SPSA-based Gradients
• Overcoming the Coherence Time Barrier in Quantum Machine Learning on Temporal Data
• Engineered dissipation to mitigate barren plateaus
• Drastic Circuit Depth Reductions with Preserved Adversarial Robustness by Approximate Encoding for Quantum Machine Learning
• Hierarchical quantum circuit representations for neural architecture search
• Preparing Valence-Bond-Solid states on noisy intermediate-scale quantum computers
• Tight and Efficient Gradient Bounds for Parameterized Quantum Circuits
• Learning shallow quantum circuits
• Effects of noise on the overparametrization of quantum neural networks
• Reduction of finite sampling noise in quantum neural networks
• Training variational quantum algorithms with random gate activation
• The Quantum Path Kernel: a Generalized Quantum Neural Tangent Kernel for Deep Quantum Machine Learning
• Evaluation of Parameterized Quantum Circuits with Cross-Resonance Pulse-Driven Entanglers
• Tackling Sampling Noise in Physical Systems for Machine Learning Applications: Fundamental Limits and Eigentasks
• Resource Saving via Ensemble Techniques for Quantum Neural Networks
• Efficient estimation of trainability for variational quantum circuits
• Quantum Convolutional Neural Networks are Effectively Classically Simulable
• A duplication-free quantum neural network for universal approximation
• Schrödinger-Heisenberg Variational Quantum Algorithms
• An introduction to variational quantum algorithms for combinatorial optimization problems
• Depth analysis of variational quantum algorithms for heat equation
• Lie-algebraic classical simulations for quantum computing
• Enabling Large-Scale and High-Precision Fluid Simulations on Near-Term Quantum Computers
• Explaining Quantum Circuits with Shapley Values: Towards Explainable Quantum Machine Learning
• Distributed quantum architecture search
• Variational Quantum Simulation of Partial Differential Equations: Applications in Colloidal Transport
• Using Differential Evolution to avoid local minima in Variational Quantum Algorithms
• A General Approach to Dropout in Quantum Neural Networks
• Classical Splitting of Parametrized Quantum Circuits
• Approximate encoding of quantum states using shallow circuits
• Isometric tensor network optimization for extensive Hamiltonians is free of barren plateaus
• Unsupervised Quantum Circuit Learning in High Energy Physics
• Simulating fluid flows with quantum computing
• Neural Quantum Embedding: Pushing the Limits of Quantum Supervised Learning
• The Unified Effect of Data Encoding, Ansatz Expressibility and Entanglement on the Trainability of HQNNs
• Connecting geometry and performance of two-qubit parameterized quantum circuits
• Solution of SAT Problems with the Adaptive-Bias Quantum Approximate Optimization Algorithm
• Quantum algorithms for scientific computing
• Atomic Quantum Technologies for Quantum Matter and Fundamental Physics Applications
• Compact quantum algorithms for time-dependent differential equations
• The quantum cost function concentration dependency on the parametrization expressivity
• Quantum Advantage Seeker with Kernels (QuASK): a software framework to speed up the research in quantum machine learning
• Quantum Energy Landscape and VQA Optimization
• Improved variational quantum eigensolver via quasi-dynamical evolution
• Barren plateaus from learning scramblers with local cost functions
• Approximately Equivariant Quantum Neural Network for $p4m$ Group Symmetries in Images
• VQC-Based Reinforcement Learning with Data Re-uploading: Performance and Trainability
• What can we learn from quantum convolutional neural networks?
• On the universality of $S_n$-equivariant $k$-body gates
• Playing Atari with Hybrid Quantum-Classical Reinforcement Learning
• Assessment of the variational quantum eigensolver: application to the Heisenberg model
• Natural parameterized quantum circuit
• Trainability and Expressivity of Hamming-Weight Preserving Quantum Circuits for Machine Learning
• Neural quantum kernels: training quantum kernels with quantum neural networks
• A blueprint for a Digital-Analog Variational Quantum Eigensolver using Rydberg atom arrays
• Semisupervised Anomaly Detection using Support Vector Regression with Quantum Kernel
• Evolution strategies: Application in hybrid quantum-classical neural networks
• Entanglement Diagnostics for Efficient Quantum Computation
• Tensor-network-assisted variational quantum algorithm
• Beyond unital noise in variational quantum algorithms: noise-induced barren plateaus and limit sets
• EHA: Entanglement-variational Hardware-efficient Ansatz for Eigensolvers
• A Benchmarking Study of Quantum Algorithms for Combinatorial Optimization
• Automatic quantum circuit encoding of a given arbitrary quantum state
• Boundary Treatment for Variational Quantum Simulations of Partial Differential Equations on Quantum Computers
• Optimizing Variational Quantum Algorithms with qBang: Efficiently Interweaving Metric and Momentum to Navigate Flat Energy Landscapes
• Mitigated barren plateaus in the time-nonlocal optimization of analog quantum-algorithm protocols
• Analyzing variational quantum landscapes with information content
• Hybrid quantum learning with data re-uploading on a small-scale superconducting quantum simulator
• The battle of clean and dirty qubits in the era of partial error correction
• Symmetry-informed transferability of optimal parameters in the Quantum Approximate Optimization Algorithm
• The role of data embedding in equivariant quantum convolutional neural networks
• Optimizing Quantum Convolutional Neural Network Architectures for Arbitrary Data Dimension
• Constrained and Vanishing Expressivity of Quantum Fourier Models
• Variational quantum eigensolver with linear depth problem-inspired ansatz for solving portfolio optimization in finance
• Estimating the randomness of quantum circuit ensembles up to 50 qubits
• Measurement-induced entanglement phase transitions in variational quantum circuits
• Large-scale simulations of Floquet physics on near-term quantum computers
• Quantum variational learning for entanglement witnessing
• Quantum Fourier Iterative Amplitude Estimation
• F-Divergences and Cost Function Locality in Generative Modelling with Quantum Circuits
• Resource frugal optimizer for quantum machine learning
• Splitting and Parallelizing of Quantum Convolutional Neural Networks for Learning Translationally Symmetric Data
• Sequential optimal selection of a single-qubit gate and its relation to barren plateau in parameterized quantum circuits
• Emergence of noise-induced barren plateaus in arbitrary layered noise models
• Characterization of variational quantum algorithms using free fermions
• Symmetry-invariant quantum machine learning force fields
• Generalization Error Bound for Quantum Machine Learning in NISQ Era -- A Survey
• Convex Optimization for Nonequilibrium Steady States on a Hybrid Quantum Processor
• Satellite image classification with neural quantum kernels
• NISQ-compatible approximate quantum algorithm for unconstrained and constrained discrete optimization
• Numerical solution of nonlinear Schrödinger equation by a hybrid pseudospectral-variational quantum algorithm
• Protocols for Trainable and Differentiable Quantum Generative Modelling
• Image Classification with Rotation-Invariant Variational Quantum Circuits
• Barren plateaus are swamped with traps
• Variational-quantum-eigensolver-inspired optimization for spin-chain work extraction
• Virtual distillation with noise dilution
• Generative model for learning quantum ensemble via optimal transport loss
• Evaluating the Resilience of Variational Quantum Algorithms to Leakage Noise
• Applicability of Measurement-based Quantum Computation towards Physically-driven Variational Quantum Eigensolver
• Variational quantum computing for quantum simulation: principles, implementations, and challenges
• Towards Efficient Ansatz Architecture for Variational Quantum Algorithms
• Adversarial Robustness Guarantees for Quantum Classifiers
• Distributed quantum machine learning via classical communication
• The dilemma of quantum neural networks
• Absence of barren plateaus and scaling of gradients in the energy optimization of isometric tensor network states
• Quantum-enhanced neural networks in the neural tangent kernel framework
• SnCQA: A hardware-efficient equivariant quantum convolutional circuit architecture
• Simulation of a feedback-based algorithm for quantum optimization for a realistic neutral atom system with an optimized small-angle controlled-phase gate
• Gradients and frequency profiles of quantum re-uploading models
• Towards Variational Quantum Algorithms for generalized linear and nonlinear transport phenomena
• On Optimizing Hyperparameters for Quantum Neural Networks
• Universal quantum computation with symmetric qubit clusters coupled to an environment
• Analyzing the quantum approximate optimization algorithm: ansätze, symmetries, and Lie algebras
• Adaptive shot allocation for fast convergence in variational quantum algorithms
• Quantum kernel learning Model constructed with small data
• Information flow in parameterized quantum circuits
• Quantum computing for genomics: conceptual challenges and practical perspectives
• Characterizing randomness in parameterized quantum circuits through expressibility and average entanglement
• A Study on Quantum Graph Neural Networks Applied to Molecular Physics
• Maximising Quantum-Computing Expressive Power through Randomised Circuits
• Sketching phase diagrams using low-depth variational quantum algorithms
• Learning the expressibility of quantum circuit ansatz using transformer
• Time-Optimal Quantum Driving by Variational Circuit Learning
• 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
• Variational Optimization for Quantum Problems using Deep Generative Networks
• QSlack: A slack-variable approach for variational quantum semi-definite programming
• Solving lattice gauge theories using the quantum Krylov algorithm and qubitization
• Characterization of overparametrization in the simulation of realistic quantum systems
• Implementing transferable annealing protocols for combinatorial optimisation on neutral atom quantum processors: a case study on smart-charging of electric vehicles
• Quantum computing quantum Monte Carlo algorithm
• Restricting to the chip architecture maintains the quantum neural network accuracy
• Single entanglement connection architecture between multi-layer bipartite Hardware Efficient Ansatz
• Parallel-in-time quantum simulation via Page and Wootters quantum time
• Enhancing Variational Quantum Circuit Training: An Improved Neural Network Approach for Barren Plateau Mitigation
• Quantum Extremal Learning
• Variational Quantum Simulation of Valence-Bond Solids
• Ground state-based quantum feature maps
• Permutation-equivariant quantum convolutional neural networks
• High-fidelity dimer excitations using quantum hardware
• Variational Microcanonical Estimator
• Quantum circuits for partial differential equations in Fourier space
• Pitfalls when tackling the exponential concentration of parameterized quantum models
• Hacking Cryptographic Protocols with Advanced Variational Quantum Attacks
• Optimized numerical gradient and Hessian estimation for variational quantum algorithms
• Universal compiling and (No-)Free-Lunch theorems for continuous variable quantum learning
• Learning to learn with an evolutionary strategy applied to variational quantum algorithms
• Towards solving the BCS Hamiltonian gap in Near-Term Quantum Computers
• Clifford circuit based heuristic optimization of fermion-to-qubit mappings
• Boson sampling enhanced quantum chemistry
• Data-Dependent Generalization Bounds for Parameterized Quantum Models Under Noise
• Random Projection using Random Quantum Circuits
• Learning Equivariant Maps with Variational Quantum Circuits
• Fourier Analysis of Variational Quantum Circuits for Supervised Learning
• Perturbative gadgets for gate-based quantum computing: Non-recursive constructions without subspace restrictions
• QKAN: quantum Kolmogorov-Arnold networks with applications in machine learning and multivariate state preparation
• Learning Fourier series with parametrized quantum circuits
• Crosstalk-Based Parameterized Quantum Circuit Approximation
• Expressibility, entangling power and quantum average causal effect for causally indefinite circuits
• A Universal Quantum Computer From Relativistic Motion
• Expressivity of deterministic quantum computation with one qubit
• Quantum Active Learning
• A Parameter-Efficient Quantum Anomaly Detection Method on a Superconducting Quantum Processor
• Graph Neural Networks for Parameterized Quantum Circuits Expressibility Estimation
• Variational quantum compiling for three-qubit gates design in quantum dots
• Evolved Quantum Boltzmann Machines
• An Efficient Gradient Sensitive Alternate Framework for VQE with Variable Ansatz
• Entanglement-informed Construction of Variational Quantum Circuits
• Barren plateaus are amplified by the dimension of qudits
• The Lie Algebra of XY-mixer Topologies and Warm Starting QAOA for Constrained Optimization
• On the Differential Topology of Expressivity of Parameterized Quantum Circuits
• Quantum Generative Training Using Rényi Divergences
• Finding eigenvectors with a quantum variational algorithm
• Mitigating the measurement overhead of ADAPT-VQE with optimised informationally complete generalised measurements
• When Quantum and Classical Models Disagree: Learning Beyond Minimum Norm Least Square
• Classical optimization with imaginary time block encoding on quantum computers: The MaxCut problem
• A general approach to quantum integration of cross sections in high-energy physics
• Improving Parameter Training for VQEs by Sequential Hamiltonian Assembly
• Generative flow-based warm start of the variational quantum eigensolver
• Quantum natural gradient with thermal-state initialization
• Variational simulation of higher-spin systems on qubit-based quantum simulators
• Lie groups for quantum complexity and barren plateau theory
• 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
• Exploiting many-body localization for scalable variational quantum simulation
• Expressiveness of Commutative Quantum Circuits: A Probabilistic Approach
• Robustness of optimized numerical estimation schemes for noisy variational quantum algorithms
• A variational quantum algorithm for entanglement quantification
• Optimisation-Free Recursive QAOA for the Binary Paint Shop Problem
• Efficient Online Quantum Circuit Learning with No Upfront Training
• A Cyclic Layerwise QAOA Training
• Multi-channel convolutional neural quantum embedding
• Integrated Encoding and Quantization to Enhance Quanvolutional Neural Networks
• Estimates of loss function concentration in noisy parametrized quantum circuits
• Global optimization in variational quantum algorithms via dynamic tunneling method
• Batched Line Search Strategy for Navigating through Barren Plateaus in Quantum Circuit Training
• Approximate Quadratization of High-Order Hamiltonians for Combinatorial Quantum Optimization
• Scaling of symmetry-restricted quantum circuits
• Quantum Phase Recognition via Quantum Attention Mechanism
• Distributed variational quantum computing with deterministic entanglement tuning
• Quantifying the Expressive Capacity of Quantum Systems: Fundamental Limits and Eigentasks
• Natural gradient and parameter estimation for quantum Boltzmann machines
• Probabilistic Representation of Commutative Quantum Circuit Models
• Learning Variational Quantum Circuit Parameters with Classical Artificial Intelligence for Quantum Phase Transition Detection
• High-expressibility Quantum Neural Networks using only classical resources
• Quantum-assisted tracer dispersion in turbulent shear flow
• Quantum generative classification with mixed states
• Mode connectivity in the QCBM loss landscape
• Gradient Scalability and Taylor Surrogation of Quantum Cost Landscapes
• Reflection-Based Adiabatic State Preparation
• Equivalence between exponential concentration in quantum machine learning kernels and barren plateaus in variational algorithms
• Hamiltonian Expressibility for Ansatz Selection in Variational Quantum Algorithms
• Schmidt quantum compressor
• Introducing Reduced-Width QNNs, an AI-inspired Ansatz Design Pattern
• Classical Neural Networks on Quantum Devices via Tensor Network Disentanglers: A Case Study in Image Classification
• Towards Compact Wavefunctions from Quantum-Selected Configuration Interaction
• The Effect of Noise on the Performance of Variational Algorithms for Quantum Chemistry
• Pulse engineering via projection of response functions
• Clustering and enhanced classification using a hybrid quantum autoencoder
• Leveraging Analog Neutral Atom Quantum Computers for Diversified Pricing in Hybrid Column Generation Frameworks
• Linear-optical protocols for mitigating and suppressing noise in bosonic systems
• Explicitly Quantum-parallel Computation by Displacements
• Improving the trainability of VQE on NISQ computers for solving portfolio optimization using convex interpolation
• Quantum feature-map learning with reduced resource overhead
• Exploiting biased noise in variational quantum models
• Double-bracket quantum algorithms for high-fidelity ground state preparation
• Performance Guarantees for Quantum Neural Estimation of Entropies
• Quantum algorithm for one-quasiparticle excitations in the thermodynamic limit via cluster-additive block diagonalization
• Moments of Quantum Channel Ensembles
• Practical Quantum Reservoir Computing in Rydberg Atom Arrays