Research map / Architectures and efficiency / Attention and transformers
QKSAN: A Quantum Kernel Self-Attention Network
Ren-Xin Zhao, Jinjing Shi, Xuelong Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use QKSAN AmHE (amplitude encoding, hardware-efficient ansatz) as best of four sub-models (ablation) | QKSAN · MNIST binary | test accuracy: 99 | Improved |
| use QKSAN AmHE on Fashion MNIST (ablation) | QKSAN · Fashion MNIST binary | test accuracy: 98.05 | Improved |
| QKSAN AmHE versus QSAN: 11 parameters and 6 layers on 4 qubits versus 9 parameters, 26 layers, 8 qubits | QKSAN AmHE · MNIST/Fashion MNIST binary | qubits and layers: 4 vs 8 qubits; 6 vs 26 layers | Improved |
| QKSAN AmHE versus QSAN convergence speed | QKSAN AmHE · MNIST binary | convergence steps: trails QSAN by 42 steps | Hurt |
| QKSAN AmHE versus QSAN convergence speed | QKSAN AmHE · Fashion MNIST binary | convergence steps: trails QSAN by 50 steps | Hurt |
| add bit flip or amplitude damping noise at probability 0.1 (ablation) | QKSAN AmHE · MNIST binary (noisy simulation) | learning accuracy: probability 0.1 | No clear effect |
| increase noise error probability beyond 0.1 (ablation) | QKSAN AmHE · MNIST binary (noisy simulation) | learning accuracy | Hurt |
| amplitude damping error versus bit flip error (ablation) | QKSAN AmHE · MNIST binary (noisy simulation) | loss function | Improved |
| use deferred measurement principle with conditional operation | QKSAN · QKSAN circuit | quantum resources: half of quantum resources | Improved |
Closest papers by meaning
- Neural Auto-designer for Enhanced Quantum Kernels (ICLR 2024 · similarity 0.87)
- Winning Lottery Tickets in Neural Networks via a Quantum-Inspired Classical Algorithm (NeurIPS 2026 · similarity 0.85)
- Optimal algorithmic complexity of inference in quantum kernel methods (NeurIPS 2026 · similarity 0.85)
- Limitations of measure-first protocols in quantum machine learning (ICML 2025 · similarity 0.85)
- Circuit Design and Efficient Simulation of Quantum Inner Product and Empirical Studies of Its Effect on Near-Term Hybrid Quantum-Classic Machine Learning (CVPR 2024 · similarity 0.84)
- Rethink the Role of Deep Learning towards Large-scale Quantum Systems (ICML 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
