Quantum End-to-End Learning for Contextual Combinatorial Optimization

2026-05-13

AI Summary

This work addresses the challenge of contextual combinatorial optimization (CCO) under uncertainty, which typically relies on NP-hard solvers, by proposing a quantum end-to-end learning framework termed QEL. QEL embeds contextual information into the Quantum Approximate Optimization Algorithm (QAOA) via a quantum state re-uploading mechanism, thereby constructing a trainable quantum policy amenable to end-to-end optimization. The framework innovatively integrates context-aware phase separators with parameterized quantum circuits, yielding an optimization-aware architecture with stability guarantees and enabling direct training driven by task-specific loss. Experimental results demonstrate that QEL achieves competitive performance while employing significantly fewer parameters than classical counterparts, highlighting its potential for realizing practical quantum advantage in industrial applications.

Abstract

Contextual combinatorial optimization (CCO) plays a critical role in decision-making under uncertainty, yet remains a significant challenge. We present Quantum End-to-End Learning (QEL), the first quantum computing-based end-to-end learning framework for CCO that leverages Quantum Approximate Optimization Algorithms. Inspired by the integration of state preparation and evolution in data re-uploading, we propose a context re-uploading phase-separator that jointly captures the complex relations among contexts, uncertain coefficients, and optimal solutions. This allows a contextual encoder to be seamlessly integrated within a quantum surrogate policy, enabling joint end-to-end training with a stationarity guarantee. Exploiting an optimization-aware structure grounded in physical principles that classical methods cannot readily leverage, our approach demonstrates practicality by directly training on task loss despite the discreteness and nonconvexity, while avoiding calls to NP-hard optimization solvers. QEL empirically achieves competitive performance while requiring substantially fewer parameters than classical benchmarks, highlighting its industrial-level potential for the future quantum era.

Problem

• Contextual Combinatorial Optimization
• Quantum Computing
• End-to-End Learning
• Decision-Making under Uncertainty
• Combinatorial Optimization

Innovation

• Quantum End-to-End Learning
• Contextual Combinatorial Optimization
• Quantum Approximate Optimization Algorithm
• Context Re-uploading
• Quantum Surrogate Policy