Inductive supervised quantum learning

Abstract

In supervised learning, an inductive learning algorithm extracts general rules from observed training instances, then the rules are applied to test instances. We show that this splitting of training and application arises naturally, in the classical setting, from a simple independence requirement with a physical interpretation of being non-signalling. Thus, two seemingly different definitions of inductive learning happen to coincide. This follows from the properties of classical information that break down in the quantum setup. We prove a quantum de Finetti theorem for quantum channels, which shows that in the quantum case, the equivalence holds in the asymptotic setting, that is, for large number of test instances. This reveals a natural analogy between classical learning protocols and their quantum counterparts, justifying a similar treatment, and allowing to inquire about standard elements in computational learning theory, such as structural risk minimization and sample complexity.

6+10 pages

References in corpus (9)

• An introduction to quantum machine learning
• Quantum-enhanced machine learning
• Quantum Circuits Architecture
• One-and-a-half quantum de Finetti theorems
• A Survey of Quantum Learning Theory
• Quantum change point
• Quantum learning of coherent states
• Generalized quantum state discrimination problems
• Finite de Finetti theorem for conditional probability distributions describing physical theories

Cited by in corpus (26)

• Quantum Machine Learning
• Efficient Learning for Deep Quantum Neural Networks
• Is quantum advantage the right goal for quantum machine learning?
• Tomography and Generative Data Modeling via Quantum Boltzmann Training
• Quantum generative adversarial network for generating discrete distribution
• Quantum machine learning for quantum anomaly detection
• Quantum autoencoders to denoise quantum data
• Quantum Neural Network States: A Brief Review of Methods and Applications
• Vulnerability of quantum classification to adversarial perturbations
• NISQ Computers: A Path to Quantum Supremacy
• Resonant Quantum Principal Component Analysis
• Machine learning \& artificial intelligence in the quantum domain
• Universal Adversarial Examples and Perturbations for Quantum Classifiers
• Quantum neural networks to simulate many-body quantum systems
• Efficient Machine Learning Representations of Surface Code with Boundaries, Defects, Domain Walls and Twists
• Unsupervised classification of quantum data
• Entanglement-guided architectures of machine learning by quantum tensor network
• QUILT: Effective Multi-Class Classification on Quantum Computers Using an Ensemble of Diverse Quantum Classifiers
• Optimal universal learning machines for quantum state discrimination
• Statistical Complexity of Quantum Learning
• Experimental demonstration of quantum learning speed-up with classical input data
• Quantum Machine Learning and its Supremacy in High Energy Physics
• Image Compression and Classification Using Qubits and Quantum Deep Learning
• Storage and retrieval of von Neumann measurements
• Quantum Edge Detection
• Measurement disturbance tradeoffs in three-qubit unsupervised quantum classification