Abstract:The advent of quantum computing holds the potential to revolutionize various fields by solving complex problems more efficiently than classical computers. Despite this promise, practical quantum advantage is hindered by current hardware limitations, notably the small number of qubits and high noise levels. In this study, we leverage adiabatic quantum computers to optimize Kolmogorov-Arnold Networks, a powerful neural network architecture for representing complex functions with minimal parameters. By modifying the network to use Bezier curves as the basis functions and formulating the optimization problem into a Quadratic Unconstrained Binary Optimization problem, we create a fixed-sized solution space, independent of the number of training samples. This strategy allows for the optimization of an entire neural network in a single training iteration in which, due to order of operations, a majority of the processing is done using a collapsed version of the training dataset. This inherently creates extremely fast training speeds, which are validated experimentally, compared to classical optimizers including Adam, Stochastic Gradient Descent, Adaptive Gradient, and simulated annealing. Additionally, we introduce a novel rapid retraining capability, enabling the network to be retrained with new data without reprocessing old samples, thus enhancing learning efficiency in dynamic environments. Experiments on retraining demonstrate a hundred times speed up using adiabatic quantum computing based optimization compared to that of the gradient descent based optimizers, with theoretical models allowing this speed up to be much larger! Our findings suggest that with further advancements in quantum hardware and algorithm optimization, quantum-optimized machine learning models could have broad applications across various domains, with initial focus on rapid retraining.

Enhanced quantum state preparation via stochastic prediction of neural network

Predictive Modelling of Quantum Process with Neural Networks

Quantum Ensemble Learning with a Programmable Superconducting Processor

Neural Predictor Based Quantum Architecture Search

Deep reinforcement learning for universal quantum state preparation via dynamic pulse control

Quantum Algorithms for Feedforward Neural Networks

Quantum State Preparation and Its Prospects in Quantum Machine Learning

Quantum Neural Network for Quantum Neural Computing

SuperEncoder: Towards Universal Neural Approximate Quantum State Preparation

A Nonstochastic Optimization Algorithm for Neural-Network Quantum States

Empowering deep neural quantum states through efficient optimization

A Non-stochastic Optimization Algorithm for Neural-network Quantum States

Optimizing quantum noise-induced reservoir computing for nonlinear and chaotic time series prediction

Experimental Quantum Stochastic Walks Simulating Associative Memory of Hopfield Neural Networks

A Quantum States Preparation Method Based on Difference-Driven Reinforcement Learning

RobustState: Boosting Fidelity of Quantum State Preparation via Noise-Aware Variational Training

Neural auto-designer for enhanced quantum kernels

Randomness-enhanced expressivity of quantum neural networks

Classifying global state preparation via deep reinforcement learning

Sparks of Quantum Advantage and Rapid Retraining in Machine Learning