Is quantum advantage the right goal for quantum machine learning?

Abstract

Machine learning is frequently listed among the most promising applications for quantum computing. This is in fact a curious choice: Today's machine learning algorithms are notoriously powerful in practice, but remain theoretically difficult to study. Quantum computing, in contrast, does not offer practical benchmarks on realistic scales, and theory is the main tool we have to judge whether it could become relevant for a problem. In this perspective we explain why it is so difficult to say something about the practical power of quantum computers for machine learning with the tools we are currently using. We argue that these challenges call for a critical debate on whether quantum advantage and the narrative of 'beating' classical machine learning should continue to dominate the literature the way it does, and highlight examples for how other perspectives in existing research provide an important alternative to the focus on advantage.

New version considerably strengthens the argument and clarifies a few points

References in corpus (7)

• An introduction to quantum machine learning
• Quantum Data Fitting
• Quantum-enhanced machine learning
• Simulating a perceptron on a quantum computer
• Focus beyond quadratic speedups for error-corrected quantum advantage
• A Feasible Approach for Automatically Differentiable Unitary Coupled-Cluster on Quantum Computers
• Perceptrons with Hebbian learning based on wave ensembles in plastic potentials

Cited by in corpus (52)

• Quantum Machine Learning: from physics to software engineering
• NISQ Computers: A Path to Quantum Supremacy
• Analytic theory for the dynamics of wide quantum neural networks
• Quantum Machine Learning for Digital Health? A Systematic Review
• Quantum phase detection generalisation from marginal quantum neural network models
• Quantum Anomaly Detection for Collider Physics
• Exponential data encoding for quantum supervised learning
• Biomarker Discovery with Quantum Neural Networks: A Case-study in CTLA4-Activation Pathways
• Microwave signal processing using an analog quantum reservoir computer
• A preprocessing perspective for quantum machine learning classification advantage using NISQ algorithms
• Quantum Convolutional Neural Networks with Interaction Layers for Classification of Classical Data
• Resource Saving via Ensemble Techniques for Quantum Neural Networks
• Explaining Quantum Circuits with Shapley Values: Towards Explainable Quantum Machine Learning
• Data re-uploading with a single qudit
• An exponentially-growing family of universal quantum circuits
• Quantum neural networks form Gaussian processes
• Quantum Advantage Seeker with Kernels (QuASK): a software framework to speed up the research in quantum machine learning
• Deep learning the hierarchy of steering measurement settings of qubit-pair states
• What can we learn from quantum convolutional neural networks?
• Quantum computing through the lens of control: A tutorial introduction
• Hyperparameter Importance of Quantum Neural Networks Across Small Datasets
• The role of data embedding in equivariant quantum convolutional neural networks
• Using (1 + 1)D Quantum Cellular Automata for Exploring Collective Effects in Large Scale Quantum Neural Networks
• Entanglement-induced provable and robust quantum learning advantages
• A Quantum Algorithm for Shapley Value Estimation
• Qudit Machine Learning
• Generalization Error Bound for Quantum Machine Learning in NISQ Era -- A Survey
• Forecasting steam mass flow in power plants using the parallel hybrid network
• Splitting and Parallelizing of Quantum Convolutional Neural Networks for Learning Translationally Symmetric Data
• Quantum Hamiltonian Embedding of Images for Data Reuploading Classifiers
• Quantum reservoir computing on random regular graphs
• Hype or Heuristic? Quantum Reinforcement Learning for Join Order Optimisation
• The Coming Decades of Quantum Simulation
• Enhanced feature encoding and classification on distributed quantum hardware
• Expressive Quantum Supervised Machine Learning using Kerr-nonlinear Parametric Oscillators
• On Optimizing Hyperparameters for Quantum Neural Networks
• Application of ZX-calculus to Quantum Architecture Search
• Dissipative quantum many-body dynamics in (1+1)D quantum cellular automata and quantum neural networks
• Characterization of overparametrization in the simulation of realistic quantum systems
• Permutation-equivariant quantum convolutional neural networks
• On the similarity of bandwidth-tuned quantum kernels and classical kernels
• Quantifying Grover speed-ups beyond asymptotic analysis
• Quantum memory and scrambling from the perspective of a classical neural network
• Quantum Neural Networks in Practice: A Comparative Study with Classical Models from Standard Data Sets to Industrial Images
• Non-linear classification capability of quantum neural networks due to emergent quantum metastability
• Quantum Curriculum Learning
• Perspectives on Utilization of Measurements in Quantum Algorithms
• Quantum Patch-Based Autoencoder for Anomaly Segmentation
• Optimal quantum reservoir learning in proximity to universality
• First-Principles Optical Descriptors and Hybrid Classical-Quantum Classification of Er-Doped CaF${}_2$
• Quantum-Enhanced Neural Exchange-Correlation Functionals
• Quantum-Inspired Weight-Constrained Neural Network: Reducing Variable Numbers by 100x Compared to Standard Neural Networks