Experimental quantum stochastic walks simulating associative memory of Hopfield neural networks

Abstract

With the increasing crossover between quantum information and machine learning, quantum simulation of neural networks has drawn unprecedentedly strong attention, especially for the simulation of associative memory in Hopfield neural networks due to their wide applications and relatively simple structures that allow for easier mapping to the quantum regime. Quantum stochastic walk, a strikingly powerful tool to analyze quantum dynamics, has been recently proposed to simulate the firing pattern and associative memory with a dependence on Hamming Distance. We successfully map the theoretical scheme into a three-dimensional photonic quantum chip and realize quantum stochastic walk evolution through well-controlled detunings of the propagation constant. We demonstrate a good match rate of the associative memory between the experimental quantum scheme and the expected result for Hopfield neural networks. The ability of quantum simulation for an important feature of a neural network, combined with the scalability of our approach through low-loss integrated chip and straightforward Hamiltonian engineering, provides a primary but steady step towards photonic artificial intelligence devices for optimization and computation tasks of greatly improved efficiencies.

arXiv comment7 pages, 4 figures

References in corpus (32)

• Quantum Machine Learning
• Deep Learning with Coherent Nanophotonic Circuits
• Quantum support vector machine for big data classification
• Machine learning phases of matter
• Quantum principal component analysis
• Universal Linear Optics
• Learning phase transitions by confusion
• The quest for a Quantum Neural Network
• Experimental boson sampling in arbitrary integrated photonic circuits
• Realization of quantum walks with negligible decoherence in waveguide lattices
• Quantum transport simulations in a programmable nanophotonic processor
• Quantum walk of a trapped ion in phase space
• A 2D Quantum Walk Simulation of Two-Particle Dynamics
• Quantum generalisation of feedforward neural networks
• Experimental Realization of Quantum Artificial Intelligence
• Experimental Implementation of the Quantum Random-Walk Algorithm
• Experimental Two-dimensional Quantum Walk on a Photonic Chip
• Entanglement-Based Machine Learning on a Quantum Computer
• Quantum Hopfield neural network
• Experimental Machine Learning of Quantum States
• Probabilistic Quantum Memories
• Quantum stochastic walks: A generalization of classical random walks and quantum walks
• Enhancing quantum transport in a photonic network using controllable decoherence
• A Separability-Entanglement Classifier via Machine Learning
• Quantum Neuron: an elementary building block for machine learning on quantum computers
• Experimental Quantum Fast Hitting on Hexagonal Graphs
• Fast Escape from Quantum Mazes in Integrated Photonics
• Tuneable quantum interference in a 3D integrated circuit
• Experimental realisation of a delayed-choice quantum walk
• Quantum walks on graphs representing the firing patterns of a quantum neural network
• Invisibility Cloak Printed on a Photonic Chip
• High-Capacity Quantum Associative Memories

Cited by in corpus (11)

• Vector Vortex Beam Emitter Embedded in a Photonic Chip
• Advances in Quantum Deep Learning: An Overview
• Generating Haar-uniform Randomness using Stochastic Quantum Walks on a Photonic Chip
• Rapid training of quantum recurrent neural networks
• Simulating Photosynthetic Energy Transport on a Photonic Network
• Efficient learning of mixed-state tomography for photonic quantum walk
• Implementation of quantum stochastic walks for function approximation, two-dimensional data classification, and sequence classification
• QSW_MPI: a framework for parallel simulation of quantum stochastic walks
• Experimental Quantum Simulation of Dynamic Localization on Curved Photonic Lattices
• State Classification via a Random-Walk-Based Quantum Neural Network
• A self-learning magnetic Hopfield neural network with intrinsic gradient descent adaption