This repository contains a complete set of implementations for the ML4SCI Quantum Machine Learning for High Energy Physics (QMLHEP) tasks.
It showcases the development of hybrid quantum-classical machine learning models using TensorFlow Quantum and Cirq, applied to physics-inspired and real-world problems.
- Task1_Quantum_Computing
- Task2_Classical_GNN
- Task3_Open_Task
- Task4_QGAN (Quantum Generative Adversarial Network)
- Task5_QGNN (Quantum Graph Neural Network)
- Task6_QRL (Quantum Representation Learning)
- Task7_EQNN (Equivariant Quantum Neural Network)
- Task8_QViT (Vision Transformer & Quantum Vision Transformer)
- Implemented multiple Quantum Machine Learning architectures from scratch
- Built QGAN, QGNN, QRL, EQNN, and Vision Transformer models
- Applied contrastive learning using SWAP test circuits
- Designed symmetry-aware quantum models (Z•2 × Z•2 equivariance)
- Explored hybrid quantum-classical optimization techniques
- Achieved strong performance on classification tasks (up to ~97% accuracy on MNIST)
- Python
- TensorFlow Quantum (TFQ)
- Cirq
- NumPy
- Hybrid Quantum-Classical Machine Learning
- Vision Transformer (MNIST): ~97% Test Accuracy
- QGAN: Improved classification beyond random baseline (AUC > 0.6)
- EQNN: Demonstrated symmetry-aware learning performance improvements
- QRL: Implemented fidelity-based contrastive learning
This work explores the intersection of:
- Quantum Machine Learning
- High Energy Physics
- Agricultural Systems Optimization
- Future domains such as:
- Post-Quantum Cryptography
- Bio-Quantum Computing
Samuel Eke
Lagos, Nigeria.
GitHub: https://github.com/JackXammie
This portfolio represents hands-on implementation of quantum machine learning models and is part of preparation for research and open-source contributions in quantum computing.