Absence of Barren Plateaus in Quantum Convolutional Neural Networks

Abstract

Quantum neural networks (QNNs) have generated excitement around the possibility of efficiently analyzing quantum data. But this excitement has been tempered by the existence of exponentially vanishing gradients, known as barren plateau landscapes, for many QNN architectures. Recently, Quantum Convolutional Neural Networks (QCNNs) have been proposed, involving a sequence of convolutional and pooling layers that reduce the number of qubits while preserving information about relevant data features. In this work we rigorously analyze the gradient scaling for the parameters in the QCNN architecture. We find that the variance of the gradient vanishes no faster than polynomially, implying that QCNNs do not exhibit barren plateaus. This provides an analytical guarantee for the trainability of randomly initialized QCNNs, which highlights QCNNs as being trainable under random initialization unlike many other QNN architectures. To derive our results we introduce a novel graph-based method to analyze expectation values over Haar-distributed unitaries, which will likely be useful in other contexts. Finally, we perform numerical simulations to verify our analytical results.

arXiv comment9 + 20 pages, 7 + 8 figures, 3 tables. Updated to published version

References in corpus (46)

• Quantum Machine Learning
• A variational eigenvalue solver on a quantum processor
• Barren plateaus in quantum neural network training landscapes
• A Quantum Approximate Optimization Algorithm
• Quantum Circuit Learning
• Quantum Convolutional Neural Networks
• An introduction to quantum machine learning
• Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits
• Evaluating analytic gradients on quantum hardware
• The power of quantum neural networks
• Noise-Induced Barren Plateaus in Variational Quantum Algorithms
• Efficient Learning for Deep Quantum Neural Networks
• The quest for a Quantum Neural Network
• Integration with respect to the Haar measure on unitary, orthogonal and symplectic group
• Connecting ansatz expressibility to gradient magnitudes and barren plateaus
• Quantum autoencoders for efficient compression of quantum data
• Optimal Quantum Circuits for General Two-Qubit Gates
• Continuous-variable quantum neural networks
• An initialization strategy for addressing barren plateaus in parametrized quantum circuits
• Quantum-assisted quantum compiling
• Layerwise learning for quantum neural networks
• Random Quantum Circuits are Approximate 2-designs
• Hierarchical quantum classifiers
• Local random quantum circuits are approximate polynomial-designs
• Variational Fast Forwarding for Quantum Simulation Beyond the Coherence Time
• Variational Quantum State Diagonalization
• Noise Resilience of Variational Quantum Compiling
• Large gradients via correlation in random parameterized quantum circuits
• Entanglement Devised Barren Plateau Mitigation
• Variational Quantum State Eigensolver
• Barren plateaus preclude learning scramblers
• Higher Order Derivatives of Quantum Neural Networks with Barren Plateaus
• Variational Quantum Fidelity Estimation
• Quantum Machine Learning in High Energy Physics
• Learning to learn with quantum neural networks via classical neural networks
• The Meta-Variational Quantum Eigensolver (Meta-VQE): Learning energy profiles of parameterized Hamiltonians for quantum simulation
• Quantum implementation of an artificial feed-forward neural network
• Symbolic integration with respect to the Haar measure on the unitary group
• Reformulation of the No-Free-Lunch Theorem for Entangled Data Sets
• Variational Consistent Histories as a Hybrid Algorithm for Quantum Foundations
• Iterative Quantum Assisted Eigensolver
• Variational Hamiltonian Diagonalization for Dynamical Quantum Simulation
• Strong bound between trace distance and Hilbert-Schmidt distance for low-rank states
• Abrupt Transitions in Variational Quantum Circuit Training
• Toward Trainability of Quantum Neural Networks
• RTNI - A symbolic integrator for Haar-random tensor networks

Cited by in corpus (81)

• Variational Quantum Algorithms
• Noisy intermediate-scale quantum (NISQ) algorithms
• The Variational Quantum Eigensolver: a review of methods and best practices
• Connecting ansatz expressibility to gradient magnitudes and barren plateaus
• Generalization in quantum machine learning from few training data
• Quantum convolutional neural network for classical data classification
• Effect of barren plateaus on gradient-free optimization
• Diagnosing Barren Plateaus with Tools from Quantum Optimal Control
• Barren Plateaus in Variational Quantum Computing
• Solving nonlinear differential equations with differentiable quantum circuits
• Trainability of Dissipative Perceptron-Based Quantum Neural Networks
• Theory of overparametrization in quantum neural networks
• Group-Invariant Quantum Machine Learning
• Generalization in Quantum Machine Learning: a Quantum Information Perspective
• Large gradients via correlation in random parameterized quantum circuits
• Equivalence of quantum barren plateaus to cost concentration and narrow gorges
• A Lie Algebraic Theory of Barren Plateaus for Deep Parameterized Quantum Circuits
• Quantum computing models for artificial neural networks
• Quantum Machine Learning: from physics to software engineering
• Theory for Equivariant Quantum Neural Networks
• Theoretical Guarantees for Permutation-Equivariant Quantum Neural Networks
• Biology and medicine in the landscape of quantum advantages
• Layer VQE: A Variational Approach for Combinatorial Optimization on Noisy Quantum Computers
• Barren plateaus in quantum tensor network optimization
• Understanding quantum machine learning also requires rethinking generalization
• On the practical usefulness of the Hardware Efficient Ansatz
• Learning quantum data with the quantum Earth Mover's distance
• Analytic theory for the dynamics of wide quantum neural networks
• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Quantum phase detection generalisation from marginal quantum neural network models
• Reflection Equivariant Quantum Neural Networks for Enhanced Image Classification
• Variational inference with a quantum computer
• RGB Image Classification with Quantum Convolutional Ansaetze
• Quantum anomaly detection in the latent space of proton collision events at the LHC
• Trainability barriers and opportunities in quantum generative modeling
• Tensor networks for quantum machine learning
• Model-Independent Learning of Quantum Phases of Matter with Quantum Convolutional Neural Networks
• Variational Quantum Anomaly Detection: Unsupervised mapping of phase diagrams on a physical quantum computer
• Long-time simulations with high fidelity on quantum hardware
• Variational Quantum Classifiers Through the Lens of the Hessian
• Dynamical simulation via quantum machine learning with provable generalization
• A semi-agnostic ansatz with variable structure for quantum machine learning
• Non-trivial symmetries in quantum landscapes and their resilience to quantum noise
• Quantum Capsule Networks
• Entangled Datasets for Quantum Machine Learning
• Classification of dynamical Lie algebras for translation-invariant 2-local spin systems in one dimension
• Engineered dissipation to mitigate barren plateaus
• Hierarchical quantum circuit representations for neural architecture search
• Drastic Circuit Depth Reductions with Preserved Adversarial Robustness by Approximate Encoding for Quantum Machine Learning
• Enhancing Detection of Topological Order by Local Error Correction
• Inference-Based Quantum Sensing
• Classical Splitting of Parametrized Quantum Circuits
• Tensor networks for quantum computing
• Barren plateaus from learning scramblers with local cost functions
• Magic of Random Matrix Product States
• Realization of a quantum neural network using repeat-until-success circuits in a superconducting quantum processor
• Automatic quantum circuit encoding of a given arbitrary quantum state
• Mitigated barren plateaus in the time-nonlocal optimization of analog quantum-algorithm protocols
• VQE Method: A Short Survey and Recent Developments
• Expressivity of Variational Quantum Machine Learning on the Boolean Cube
• Measurement-induced entanglement phase transitions in variational quantum circuits
• Coarse grained intermolecular interactions on quantum processors
• Quantum Goemans-Williamson Algorithm with the Hadamard Test and Approximate Amplitude Constraints
• Optimal storage capacity of quantum Hopfield neural networks
• Resource frugal optimizer for quantum machine learning
• Laziness, Barren Plateau, and Noise in Machine Learning
• Qsun: an open-source platform towards practical quantum machine learning applications
• Quantum algorithms for group convolution, cross-correlation, and equivariant transformations
• Adaptive shot allocation for fast convergence in variational quantum algorithms
• Efficient Learning for Linear Properties of Bounded-Gate Quantum Circuits
• Application of Quantum Tensor Networks for Protein Classification
• Time-Optimal Quantum Driving by Variational Circuit Learning
• Using gradient-based algorithms to determine ground state energies on a quantum computer
• Exploring Ground States of Fermi-Hubbard Model on Honeycomb Lattices with Counterdiabaticity
• Quantum Extremal Learning
• Universal compiling and (No-)Free-Lunch theorems for continuous variable quantum learning
• Hybrid Quantum Classical Graph Neural Networks for Particle Track Reconstruction
• qLEET: Visualizing Loss Landscapes, Expressibility, Entangling Power and Training Trajectories for Parameterized Quantum Circuits
• Mitigating Noise-Induced Gradient Vanishing in Variational Quantum Algorithm Training
• Optimizing Quantum Variational Circuits with Deep Reinforcement Learning
• Quantum Machine Learning with SQUID