Module quantum_meta_learning 

Quantum Meta-Learning

This module implements meta-learning algorithms for quantum machine learning, enabling rapid adaptation to new tasks with minimal training data.

Theoretical Background

Quantum meta-learning extends classical meta-learning (MAML, Reptile) to quantum neural networks. The goal is to learn an initialization of quantum circuit parameters that can quickly adapt to new tasks through fine-tuning.

Key Algorithms

• Quantum MAML: Model-Agnostic Meta-Learning for quantum circuits
• Quantum Reptile: First-order approximation of MAML
• Quantum ProtoNets: Prototype networks using quantum metric learning
• Quantum Matching Networks: Attention-based few-shot learning

Applications

• Few-shot quantum classification
• Fast quantum state tomography
• Adaptive quantum control
• Quantum drug discovery with limited data

References

• “Meta-Learning for Quantum Neural Networks”
• “Few-Shot Learning with Quantum Classifiers”
• “Quantum Model-Agnostic Meta-Learning”

Structs

QuantumMAML

Quantum MAML (Model-Agnostic Meta-Learning)

QuantumMetaCircuit

Quantum circuit for meta-learning

QuantumMetaLearningConfig

Configuration for quantum meta-learning

QuantumReptile

Quantum Reptile (simpler first-order MAML)

QuantumTask

Quantum task for meta-learning