Quantum Capsule Networks

Abstract

Capsule networks, which incorporate the paradigms of connectionism and symbolism, have brought fresh insights into artificial intelligence. The capsule, as the building block of capsule networks, is a group of neurons represented by a vector to encode different features of an entity. The information is extracted hierarchically through capsule layers via routing algorithms. Here, we introduce a quantum capsule network (dubbed QCapsNet) together with an efficient quantum dynamic routing algorithm. To benchmark the performance of the QCapsNet, we carry out extensive numerical simulations on the classification of handwritten digits and symmetry-protected topological phases, and show that the QCapsNet can achieve an enhanced accuracy and outperform conventional quantum classifiers evidently. We further unpack the output capsule state and find that a particular subspace may correspond to a human-understandable feature of the input data, which indicates the potential explainability of such networks. Our work reveals an intriguing prospect of quantum capsule networks in quantum machine learning, which may provide a valuable guide towards explainable quantum artificial intelligence.

arXiv comment7 pages (main text) + 8 pages (supplementary information), 8 figures

References in corpus (66)

• TensorFlow: A system for large-scale machine learning
• Quantum Computing in the NISQ era and beyond
• Supplementary information for "Quantum supremacy using a programmable superconducting processor"
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
• The Quantum Internet
• Quantum Machine Learning
• Variational Quantum Algorithms
• Quantum computational advantage using photons
• Barren plateaus in quantum neural network training landscapes
• Quantum support vector machine for big data classification
• Quantum Convolutional Neural Networks
• Quantum Amplitude Amplification and Estimation
• Quantum principal component analysis
• Quantum fingerprinting
• Strong quantum computational advantage using a superconducting quantum processor
• Parameterized quantum circuits as machine learning models
• Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits
• Quantum random access memory
• Noise-Induced Barren Plateaus in Variational Quantum Algorithms
• Efficient Learning for Deep Quantum Neural Networks
• Efficient quantum algorithms for simulating sparse Hamiltonians
• Quantum generative adversarial learning
• Quantum algorithms for supervised and unsupervised machine learning
• Benchmarking an 11-qubit quantum computer
• Quantum generative adversarial networks
• Phase-Programmable Gaussian Boson Sampling Using Stimulated Squeezed Light
• An initialization strategy for addressing barren plateaus in parametrized quantum circuits
• Observation of multi-component atomic Schrödinger cat states of up to 20 qubits
• Absence of Barren Plateaus in Quantum Convolutional Neural Networks
• Quantum Boltzmann Machine
• Hierarchical quantum classifiers
• Architectures for a quantum random access memory
• Quantum walks on a programmable two-dimensional 62-qubit superconducting processor
• Hybrid Quantum-Classical Convolutional Neural Networks
• Effect of barren plateaus on gradient-free optimization
• Trainability of Dissipative Perceptron-Based Quantum Neural Networks
• Hardware-efficient quantum random access memory with hybrid quantum acoustic systems
• Quantum generative adversarial learning in a superconducting quantum circuit
• Cross-Platform Verification of Intermediate Scale Quantum Devices
• Machine learning meets quantum physics
• Equivalence of quantum barren plateaus to cost concentration and narrow gorges
• Circuit-Based Quantum Random Access Memory for Classical Data
• Entanglement Devised Barren Plateau Mitigation
• Quantum principal component analysis only achieves an exponential speedup because of its state preparation assumptions
• Quantum Adversarial Machine Learning
• Statistical mechanics of the Cluster-Ising model
• Barren plateaus preclude learning scramblers
• Higher Order Derivatives of Quantum Neural Networks with Barren Plateaus
• On barren plateaus and cost function locality in variational quantum algorithms
• Quantum federated learning through blind quantum computing
• Experimental quantum adversarial learning with programmable superconducting qubits
• Recent advances for quantum classifiers
• Vulnerability of quantum classification to adversarial perturbations
• Quantum Algorithms for Deep Convolutional Neural Networks
• Representation Learning via Quantum Neural Tangent Kernels
• Analyzing the barren plateau phenomenon in training quantum neural networks with the ZX-calculus
• Learning quantum data with the quantum Earth Mover's distance
• Quantum Wasserstein Generative Adversarial Networks
• Markovian Quantum Neuroevolution for Machine Learning
• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Conceptual understanding through efficient inverse-design of quantum optical experiments
• ZhuSuan: A Library for Bayesian Deep Learning
• Distributed quantum inner product estimation
• Representation of binary classification trees with binary features by quantum circuits
• Quantum Expectation-Maximization Algorithm
• Quantum Expectation-Maximization for Gaussian Mixture Models

Cited by in corpus (8)

• Harnessing Quantum Extreme Learning Machines for image classification
• Distributed quantum architecture search
• Hybrid quantum-classical unsupervised data clustering based on the self-organizing feature map
• Universal adversarial perturbations for multiple classification tasks with quantum classifiers
• Enhancing Variational Quantum Circuit Training: An Improved Neural Network Approach for Barren Plateau Mitigation
• Quantum Active Learning
• Quantum approximated cloning-assisted density matrix exponentiation
• The Dual Role of Low-Weight Pauli Propagation: A Flawed Simulator but a Powerful Initializer for Variational Quantum Algorithms