quantum-kernels

Quantum kernel estimation with backend-matched IBM noise modeling, plus reproducible “Wigner’s friend” branch-transfer coherence-witness experiments executed on superconducting quantum hardware.

Verification harness for quantum ML. A reproducible lab for stress-testing quantum models where predictive accuracy, identifiability, curvature, and robustness under noise can diverge.

Controlled kernel-swap comparison of quantum and classical kernels under distribution shift (EMBER, UNSW-NB15, ToN-IoT). Kernel geometry governs OOD robustness: 72 settings x 15 runs, GP + SVC classifiers, mechanism analysis. Fully reproducible artifact.

COMP47950: compare classical ML, simulated QML (VQC & QSVM), and IBM Quantum inference-only evaluation on Nim classification.

Entangled group-invariant quantum kernels for visual symmetry and anomaly experiments.

Quantum LeJEPA: applying LeJEPA self-supervised learning to quantum feature maps, and the second-moment operator M2 that predicts quantum-kernel trainability without training. Personal research.

Hands-on quantum computing for machine learning, built from scratch in NumPy then bridged to PennyLane. 6-week course: qubits & gates, entanglement & CHSH, quantum algorithms (Deutsch–Jozsa, Grover, QFT), variational circuits & VQE, quantum ML classifiers, and quantum kernels.

Reproducibility code and summary artifacts for 'Exact Transfer-Matrix Evaluation Reveals Operating-Point Limits and Opportunities for Fidelity Quantum Kernels in Biomedical Abstract Screening' (QSVC for biomedical abstract screening).