Data & Analytics

Quantum annealing-based feature selection for interpretable AI in Convolutional Neural Networks. Uses constrained optimization to select most important feature maps contributing to predictions, providing explainable AI with improved class disentanglement. Use when implementing XAI for CNNs, quantum annealing feature selection, or model interpretation via quantum computing.

Quantum adversarial defense methodology using quantum autoencoders for protecting quantum classifiers against adversarial perturbations. Covers quantum autoencoder purification, adversarial training-free defense frameworks, confidence metrics for adversarial sample detection, and evaluation of variational quantum classifiers under attack. Use when defending QML models, analyzing quantum adversarial robustness, implementing purification-based defenses, or studying adversarial attacks on variat...

Quantum Machine Learning methodology for 6G edge network adaptive communication and model aggregation in V2X systems. Combines quantum ML with edge computing for efficient vehicular communication, model collaboration, and generalization. Use when: (1) designing 6G quantum-enhanced networks, (2) V2X communication optimization, (3) edge AI model aggregation, (4) quantum ML for communication systems, (5) adaptive quantum edge networks.

Quantized time statistics methodology for quantum walks under weak rank-K measurements using topological winding numbers

Quantum measurement return statistics methodology analyzing quantized mean return time under strong and weak monitoring. Connects winding number topology with statistical properties of quantum state recurrence.

Mutation testing framework for Quantum Neural Networks (QNNs) based on the QuanForge methodology (arXiv:2604.20706). Use this skill when testing QNN robustness, analyzing quantum circuit vulnerabilities, performing mutation testing on quantum ML models, localizing weak regions in quantum circuits, or comparing QNN test suites. Also triggered by keywords: quantum testing, QNN testing, mutation testing, 量子测试, 量子神经网络测试.

SPATE: Spiking-Phase Adaptive Temporal Encoding for Quantum Machine Learning. Bridges neuromorphic computing with QML via spike-based temporal encoding into phase-encoded qubits. Use when: spiking quantum encoding, QML temporal encoding, spike encoding quantum, neuromorphic quantum computing, temporal data for QML, 脉冲量子编码.

Quantum Machine Learning model testing and robustness analysis methodology. Covers mutation testing for QNN circuits, accuracy/robustness evaluation of Variational Quantum Circuits (VQCs), and practical considerations for deploying QML models on NISQ-era quantum hardware. Use when: (1) testing quantum neural network implementations for correctness, (2) evaluating QML model robustness against circuit faults and noise, (3) designing test suites for parametrized quantum circuits, (4) analyzing V...

Design framework-agnostic quantum machine learning (QML) systems using the Model-Agnostic Learning System (MALS) paradigm. Extracts QML models from any framework (PennyLane, Qiskit, TensorFlow Quantum, etc.) into portable representations with auto-validation and cross-framework compatibility testing.

Quantum Machine Learning expressivity separation methodology. Based on Anschuetz & Gao (Quantum 10, 1976, 2026). Provides framework for constructing efficiently trainable QNNs with provable polynomial memory separations over classical neural networks. Use when: (1) designing QNN architectures with provable quantum advantage, (2) analyzing expressivity vs trainability trade-offs, (3) implementing quantum contextuality as computational resource, (4) comparing quantum vs classical sequence model...

Quantum Machine Learning with Equilibrium Propagation for medical image analysis. Energy-based training without backpropagation using Variational Quantum Circuits (VQCs) for resource-constrained quantum hardware. Use when: analyzing blood cells, leukemia detection, medical imaging with QML, energy-based quantum training, backprop-free quantum networks, or evaluating QML feasibility on NISQ devices.

Unsupervised clustering via steady-state quantum transport in open quantum networks (GKSL master equation). Encodes data as input states and infers cluster assignments from terminal current observables - no full state tomography required. Use when: quantum clustering, GKSL transport, analog quantum ML, open quantum network clustering, Qlustering algorithm, steady-state quantum transport clustering, tomography-free quantum learning, quantum unsupervised learning, algorithm-hardware co-design c...

QDiffusion-TS - First quantum generative diffusion model for time series synthesis with real quantum hardware validation on IQM processor

Parallel multi-circuit quantum feature fusion methodology for medical image classification. Use when: (1) building hybrid quantum-classical CNN architectures for biomedical image classification, (2) comparing quantum vs classical models with statistical rigor (Wilcoxon signed-rank test, Cohen's d effect size), (3) designing parallel quantum encoding circuits (amplitude + angle encoding simultaneously), (4) parameter-matched fairness evaluation for QML vs classical baselines. Covers QCNN archi...

多目标量子工作流优化方法论。系统化选择 NISQ 设备上的编译策略、噪声抑制和误差缓解方案。基于 QBalance 框架，涵盖加权目标函数、非支配选择规则、生存乘积误差代理、贝叶斯候选排序和分布诊断。Activation: qbalance, quantum workflow, quantum compilation optimization, NISQ error mitigation, quantum noise suppression, multi-objective quantum strategy.

Quantum autoencoder (QAE) for compression-driven anomaly detection in brain MRI data - angle encoding into quantum states, variational encoder-decoder with trash qubits, achieving 0.95 slice-level ROC-AUC.

Quantum Algorithm for Distributed Reduction of Entanglements (QADR) — hybrid quantum-classical ML framework that decomposes global VQCs into localized sub-circuits within causal light cones. Reduces classical simulation memory from O(2^n) to O(2^d) while mitigating barren plateaus. arXiv:2606.01291

Pulse-level Quantum Fourier Models (QFMs) for quantum machine learning. Use when: (1) implementing variational quantum algorithms at the pulse/hardware level, (2) optimizing QFM training landscapes, (3) designing pulse-parameterized quantum circuits, (4) analyzing expressibility and Fourier coefficient correlation of quantum models, (5) replacing gate-level parameterization with pulse-level control. Activation: pulse-level quantum computing, quantum Fourier models, QFM training optimization, ...

Pulse-level quantum computing skill — design, optimize, and analyze pulse-level variational quantum algorithms beyond the gate abstraction. Covers pulse parameterization, expressibility, Fourier coefficient correlation (FCC), composite gate sub-angle decomposition, and training landscape optimization. Use when: pulse-level quantum computing, variational quantum algorithms, quantum machine learning at pulse level, Fourier quantum models, QFM optimization, pulse parameterization, quantum compil...

Pulse-level Quantum Fourier Models (QFMs) for quantum machine learning. Optimizes variational quantum algorithms by using pulse parameters instead of gate-level angles, providing higher-dimensional escape routes in the optimization landscape. Use when: designing pulse-level quantum circuits, optimizing QFM training, improving variational quantum algorithm convergence, working with quantum machine learning expressibility and Fourier coefficient correlation, or replacing gate-level parameteriza...

精神病早期阶段脑动力学临界性scaling偏差研究方法论。结合重整化群(RG)框架与多种scaling分析方法，揭示临界 regime内的动力学重组而非临界性丧失。

Predict-Reuse-Repair (PRR) runtime for accelerating dynamic sparse attention in long-context LLM decoding. Speculates attention over predicted KV blocks while selection is in flight, then incrementally repairs missed blocks. Reduces per-token decoding latency up to 40%.

**arXiv ID:** 2411.00222 **Authors:** Ehsan Ganjidoost, Jeff Orchard **Published:** 2024-10-31T21:38:05Z **Abstract:** An adversarial example is a modified input image designed to cause a Machine Learning (ML) model to make a mistake; these perturbations are often invisible or subtle to human observers and highlight vulnerabilities in a model's ability to generalize from its training data. Several adversarial attacks can create such examples, each with a different perspective, effectiveness, ...

**arXiv ID:** 2207.04884 **Authors:** Chikako Dozono, Mina Aragaki, Hana Hebishima, Shin-ichi Inage **Published:** 2022-06-23T06:47:32Z **Abstract:** This paper aims at proposing a new machine learning for classification problems. The classification problem has a wide range of applications, and there are many approaches such as decision trees, neural networks, and Bayesian nets. In this paper, we focus on the action of neurons in the brain, especially the EPSP/IPSP cancellation between excita...