Noise-Robust Hybrid Quantum Neural Networks (HQNNs)
This repository contains the complete experimental codebase for my Master’s thesis, which investigates the reliability, robustness, and practical limitations of hybrid quantum–classical neural networks (HQNNs) in the NISQ era.
The project consists of a 13-demo experimental ecosystem implemented across Qiskit, Cirq, and PennyLane, combining:
- hybrid quantum learning (HQNNs, SPSA)
- variational algorithms (VQE, QAOA)
- quantum kernel methods (QSVM)
- cross-framework correctness validation
- noise robustness benchmarking
- applied case studies (healthcare, energy, cybersecurity)
📌 This repository is intended as a research-grade system.
Demonstration Ecosystem (13 Demos)
This repository includes thirteen experimental demonstrations used in the thesis. Each demo validates a different part of the hybrid architecture.
Core Demos
HQNN Toy Classifier (Qiskit)
VQE Energy Minimization (PennyLane)
QAOA MaxCut (Cirq)
QSVM Anomaly Detection (Qiskit ML)
Noise-Robust HQNN (Qiskit)
Cross-Framework Noise Benchmark (Qiskit/Cirq/PennyLane)
Endianness Parity Consistency (Qiskit/Cirq/PennyLane)
Hybrid HQNN Training Loop with SPSA
Industry-Inspired Demos
Medical Risk Classification (HQNN vs Classical)
Energy Grid Optimization using QAOA (Cirq)
Cybersecurity Anomaly Detection (QSVM + HQNN)
HQNN Explainability and Sensitivity Analysis
Cross-Noise Robustness Heatmap (Qiskit/Cirq/PennyLane)
Full descriptions of each demonstration are provided in the Demo_Descriptions.pdf document included in the thesis package.
How to Run a Demo
Each demo is contained inside the demos/ folder.
For example:
cd demos/core
python demo01_hqnn_toy_classifier.py
Or for an industry demo:
cd demos/industry
python demo10_energy_qaoa.py
Some demos require additional dependencies. Instructions will be added as diagrams and detailed documentation are integrated.
Environment Setup
A simple environment setup is provided:
python3 -m venv env
source env/bin/activate
pip install -r requirements.txt
Pseudocode
All pseudocode files for the thirteen demos are included in the thesis submission ZIP under:
4_Pseudocode/
They are not duplicated in this repository to avoid clutter and maintain a clean structure.
Documentation
Complete documentation is contained in:
Technical Manuscript (PDF)
Thesis Pre-Draft (~300 pages)
Demo Descriptions (PDF)
Slide Deck (50 slides)
These are included in the thesis submission package sent to the advisor.
Status
The codebase is complete. The next update will include:
architecture diagrams
circuit diagrams
BPMN-style workflow diagrams
demonstration figures and plots
Contact
For inquiries or questions about the project:
GitHub: https://github.com/jeragilo/