Module 21 lesson path
Module 2
AI for Quantum Hardware and Optimization
Explains how classical AI supports quantum routing, constrained optimization, graph shrinking, and realistic problem reformulation.
Learning goals
- Understand why classical AI often enables rather than replaces quantum computation.
- Explain routing, graph shrinking, and augmented Lagrangian methods in hardware-aware terms.
- Connect combinatorial reformulation to qubit scarcity.
Source highlights
- Dimensionality Reduction: Learning-Based Graph Shrinking
- Dynamic Penalty Tuning via Soft Actor-Critic (SAC)
- Overcoming Topological Constraints through Nested Qubit Routing
Lessons
Module lessons and study paths
Routing, Graph Shrinking, and Logistics under Hardware Constraints
Connects logistics QUBO reformulation, learning-based graph shrinking, SAC-tuned augmented Lagrangian methods, and nested routing into a single hardware-aware quantum logistics pipeline.
- Quantum logistics becomes tractable only after classical preprocessing compresses the instance while preserving feasibility.
- Graph shrinking is modeled as an MDP with a GNN policy so the reduced graph still respects combinatorial constraints.
- Routing remains a physical compilation bottleneck even after reformulation, so post-routing depth control still decides executable value.