Quantum optical classifier with superexponential speedup
arXiv:2404.15266 · doi:10.1038/s42005-025-02020-5
Abstract
Classification is a central task in deep learning algorithms. Usually, images are first captured and then processed by a sequence of operations, of which the artificial neuron represents one of the fundamental units. This paradigm requires significant resources that scale (at least) linearly in the image resolution, both in terms of photons and computational operations. Here, we present a quantum optical pattern recognition method for binary classification tasks. It classifies objects without reconstructing their images, using the rate of two-photon coincidences at the output of a Hong-Ou-Mandel interferometer, where both the input and the classifier parameters are encoded into single-photon states. Our method exhibits the behaviour of a classical neuron of unit depth. Once trained, it shows a constant complexity in the number of computational operations and photons required by a single classification. This is a superexponential advantage over a classical artificial neuron.
14 pages, 6 figures; [v2] Additional simulations, figures and overall improvements
References in corpus (22)
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- Variational Quantum Algorithms
- All-Optical Machine Learning Using Diffractive Deep Neural Networks
- Photonics for artificial intelligence and neuromorphic computing
- Quantum algorithms for supervised and unsupervised machine learning
- All Optical Neural Network with Nonlinear Activation Functions
- The physics of optical computing
- Continuous-variable quantum neural networks
- An Artificial Neuron Implemented on an Actual Quantum Processor
- Entanglement-Based Machine Learning on a Quantum Computer
- The SWAP test and the Hong-Ou-Mandel effect are equivalent
- Computational Imaging Without a Computer: Seeing Through Random Diffusers at the Speed of Light
- Quantum machine learning for image classification
- An Optical Frontend for a Convolutional Neural Network
- High-dimensional two-photon interference effects in spatial modes
- Quantum computing model of an artificial neuron with continuously valued input data
- Near-100 % two-photon-like coincidence-visibility dip with classical light and the role of complementarity
- All-Optical Image Identification with Programmable Matrix Transformation
- Quantum verification of NP problems with single photons and linear optics
- Continuous Variable Quantum Perceptron
- Quantum Kernel Evaluation via Hong-Ou-Mandel Interference
- A Kerr kernel quantum learning machine