Universal discriminative quantum neural networks

Abstract

Quantum mechanics fundamentally forbids deterministic discrimination of quantum states and processes. However, the ability to optimally distinguish various classes of quantum data is an important primitive in quantum information science. In this work, we train near-term quantum circuits to classify data represented by non-orthogonal quantum probability distributions using the Adam stochastic optimization algorithm. This is achieved by iterative interactions of a classical device with a quantum processor to discover the parameters of an unknown non-unitary quantum circuit. This circuit learns to simulates the unknown structure of a generalized quantum measurement, or Positive-Operator-Value-Measure (POVM), that is required to optimally distinguish possible distributions of quantum inputs. Notably we use universal circuit topologies, with a theoretically motivated circuit design, which guarantees that our circuits can in principle learn to perform arbitrary input-output mappings. Our numerical simulations show that shallow quantum circuits could be trained to discriminate among various pure and mixed quantum states exhibiting a trade-off between minimizing erroneous and inconclusive outcomes with comparable performance to theoretically optimal POVMs. We train the circuit on different classes of quantum data and evaluate the generalization error on unseen mixed quantum states. This generalization power hence distinguishes our work from standard circuit optimization and provides an example of quantum machine learning for a task that has inherently no classical analogue.

arXiv comment19 pages, 10 figures

References in corpus (53)

• Adam: A Method for Stochastic Optimization
• Topological insulators and superconductors
• The Quantum Internet
• Quantum Machine Learning
• Quantum sensing
• Advances in Quantum Metrology
• Quantum algorithm for solving linear systems of equations
• Barren plateaus in quantum neural network training landscapes
• Quantum support vector machine for big data classification
• Quantum Circuit Learning
• Quantum Convolutional Neural Networks
• Quantum principal component analysis
• Circuit-centric quantum classifiers
• Exact solutions to the nonlinear dynamics of learning in deep linear neural networks
• Ground-to-satellite quantum teleportation
• Synthesis of Quantum Logic Circuits
• Quantum algorithm for systems of linear equations with exponentially improved dependence on precision
• Scalable Designs for Quasiparticle-Poisoning-Protected Topological Quantum Computation with Majorana Zero Modes
• Quantum generative adversarial learning
• Quantum algorithms for supervised and unsupervised machine learning
• Quantum machine learning: a classical perspective
• Quantum autoencoders for efficient compression of quantum data
• Microwave Quantum Illumination
• Quantum Illumination at the Microwave Wavelengths
• Probabilistic cloning and identification of linearly independent quantum states
• Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
• Quantum State Discrimination
• Quantum-assisted quantum compiling
• Quantum Boltzmann Machine
• Hierarchical quantum classifiers
• Quantum generalisation of feedforward neural networks
• A quantum linear system algorithm for dense matrices
• Unitary Evolution Recurrent Neural Networks
• Quantum Circuits for Isometries
• Variational algorithms for linear algebra
• Designing Optimal Quantum Detectors Via Semidefinite Programming
• An Adaptive Optimizer for Measurement-Frugal Variational Algorithms
• Optical realization of optimal unambiguous discrimination for pure and mixed quantum states
• Optical codeword demodulation with error rates below standard quantum limit using a conditional nulling receiver
• Eigenvalues of the Hessian in Deep Learning: Singularity and Beyond
• A quantum algorithm to train neural networks using low-depth circuits
• Reduction Theorems for Optimal Unambiguous State Discrimination of Density Matrices
• Learning hard quantum distributions with variational autoencoders
• Quantum gate learning in engineered qubit networks: Toffoli gate with always-on interactions
• Variable time amplitude amplification and a faster quantum algorithm for solving systems of linear equations
• Quantum linear systems algorithms: a primer
• Smoothed Analysis of Algorithms: Why the Simplex Algorithm Usually Takes Polynomial Time
• Quantum Circuits for Quantum Channels
• Supervised learning of time-independent Hamiltonians for gate design
• Quantum Inspired Training for Boltzmann Machines
• Quantum algorithm for optical template recognition with noise filtering
• Quantum-classical truncated Newton method for high-dimensional energy landscapes
• Saddle-free Hessian-free Optimization

Cited by in corpus (35)

• The effect of data encoding on the expressive power of variational quantum machine learning models
• Layerwise learning for quantum neural networks
• General parameter-shift rules for quantum gradients
• PennyLane: Automatic differentiation of hybrid quantum-classical computations
• Barren Plateaus in Variational Quantum Computing
• Structure optimization for parameterized quantum circuits
• Edge AI: A Taxonomy, Systematic Review and Future Directions
• Quantum Machine Learning for Chemistry and Physics
• Adversarial quantum circuit learning for pure state approximation
• Understanding quantum machine learning also requires rethinking generalization
• Realization of a quantum autoencoder for lossless compression of quantum data
• Dynamical mean field theory algorithm and experiment on quantum computers
• Quantum Computing: An Overview Across the System Stack
• RGB Image Classification with Quantum Convolutional Ansaetze
• Optimal Provable Robustness of Quantum Classification via Quantum Hypothesis Testing
• Artificial Intelligence for Quantum Computing
• Quantum Architecture Search with Meta-learning
• Alleviating Barren Plateaus in Parameterized Quantum Machine Learning Circuits: Investigating Advanced Parameter Initialization Strategies
• Quantum State Discrimination Using Noisy Quantum Neural Networks
• Resource Saving via Ensemble Techniques for Quantum Neural Networks
• Experimental quantum state discrimination using the optimal fixed rate of inconclusive outcomes strategy
• Introduction to UniversalQCompiler
• Variational quantum state discriminator for supervised machine learning
• Quantum-machine-learning channel discrimination
• Energy-dependent barren plateau in bosonic variational quantum circuits
• Computing Graph Edit Distance with Algorithms on Quantum Devices
• Hardware-efficient entangled measurements for variational quantum algorithms
• QFCNN: Quantum Fourier Convolutional Neural Network
• State Classification via a Random-Walk-Based Quantum Neural Network
• Training quantum measurement devices to discriminate unknown non-orthogonal quantum states
• Quantum Machine Learning For Classical Data
• A hybrid quantum-classical neural network with deep residual learning
• Quantum-data-driven dynamical transition in quantum learning
• Variational-toolbox-based separability detection of multiqubit states
• Fast suppression of classification error in variational quantum circuits