Quantum End-to-End Learning for Contextual Combinatorial Optimization

TL;DR

This paper introduces Quantum End-to-End Learning (QEL), a novel quantum computing framework for contextual combinatorial optimization that leverages quantum algorithms and joint training to improve efficiency and performance.

Contribution

QEL is the first quantum-based end-to-end learning framework for CCO, integrating quantum algorithms with a context re-uploading mechanism for joint training.

Findings

QEL achieves competitive performance with fewer parameters than classical methods.

QEL avoids NP-hard optimization calls by directly training on task loss.

QEL demonstrates practical potential for industrial applications in the quantum era.

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.