Representation Learning via Quantum Neural Tangent Kernels
arXiv:2111.04225 · doi:10.1103/PRXQuantum.3.030323
Abstract
Variational quantum circuits are used in quantum machine learning and variational quantum simulation tasks. Designing good variational circuits or predicting how well they perform for given learning or optimization tasks is still unclear. Here we discuss these problems, analyzing variational quantum circuits using the theory of neural tangent kernels. We define quantum neural tangent kernels, and derive dynamical equations for their associated loss function in optimization and learning tasks. We analytically solve the dynamics in the frozen limit, or lazy training regime, where variational angles change slowly and a linear perturbation is good enough. We extend the analysis to a dynamical setting, including quadratic corrections in the variational angles. We then consider hybrid quantum-classical architecture and define a large-width limit for hybrid kernels, showing that a hybrid quantum-classical neural network can be approximately Gaussian. The results presented here show limits for which analytical understandings of the training dynamics for variational quantum circuits, used for quantum machine learning and optimization problems, are possible. These analytical results are supported by numerical simulations of quantum machine learning experiments.
v1,v2: 40=11+29 pages, many figures. v3: clearing typos, responding feedbacks, and minor modifications
References in corpus (4)
Cited by in corpus (36)
- Generalization in quantum machine learning from few training data
- A comprehensive review of Quantum Machine Learning: from NISQ to Fault Tolerance
- Theoretical Guarantees for Permutation-Equivariant Quantum Neural Networks
- Quantum-centric Supercomputing for Materials Science: A Perspective on Challenges and Future Directions
- Analytic theory for the dynamics of wide quantum neural networks
- Quantum Phase Recognition via Quantum Kernel Methods
- Quantum Capsule Networks
- Effects of noise on the overparametrization of quantum neural networks
- Let Quantum Neural Networks Choose Their Own Frequencies
- The Quantum Path Kernel: a Generalized Quantum Neural Tangent Kernel for Deep Quantum Machine Learning
- Resource Saving via Ensemble Techniques for Quantum Neural Networks
- Pulse-efficient quantum machine learning
- Stochastic noise can be helpful for variational quantum algorithms
- Quantum neural networks form Gaussian processes
- GSQAS: Graph Self-supervised Quantum Architecture Search
- Embedding Learning in Hybrid Quantum-Classical Neural Networks
- Expressivity of Variational Quantum Machine Learning on the Boolean Cube
- Estimating the randomness of quantum circuit ensembles up to 50 qubits
- Measurement-based infused circuits for variational quantum eigensolvers
- Explainable Representation Learning of Small Quantum States
- Quantum Lazy Training
- Energy-dependent barren plateau in bosonic variational quantum circuits
- Laziness, Barren Plateau, and Noise in Machine Learning
- SnCQA: A hardware-efficient equivariant quantum convolutional circuit architecture
- Characterization of overparametrization in the simulation of realistic quantum systems
- Trained quantum neural networks are Gaussian processes
- Quantitative convergence of trained quantum neural networks to a Gaussian process
- Generalization Bounds in Hybrid Quantum-Classical Machine Learning Models
- SU(d)-Symmetric Random Unitaries: Quantum Scrambling, Error Correction, and Machine Learning
- Benchmarking of quantum fidelity kernels for Gaussian process regression
- Dynamical transition in controllable quantum neural networks with large depth
- Quantum-data-driven dynamical transition in quantum learning
- Double descent in quantum kernel methods
- Gradient Scalability and Taylor Surrogation of Quantum Cost Landscapes
- Efficient classical computation of the neural tangent kernel of quantum neural networks
- Kernel Reboot: Breaking the Boundaries of Neural Tangent Kernels for Neural Fields