QML Class Spring 2024 - Quantum Classifier Project

Project Overview

This project involves simulating a quantum classifier using a variational quantum circuit for binary classification problems. It is divided into three main parts, each contributing to the total project credits.

Part 1: Quantum Classifier for 3-input Parity Problem (6/10 Credits)

Requirements:

• Objective: Build a quantum classifier for the binary classification parity problem with 3 inputs.
• Data: Provided train and test datasets (classA_train.dat, classB_train.dat, classA_test.dat, classB_test.dat).
• Data Preprocessing: Allowed to shift the data or perform simple preprocessing like subtracting 0.5 from all data.

Deliverables:

1. Circuit Design: Design the quantum circuit including measurements (recommendation: do not exceed 4 qubits).
2. Cost Function: Define the cost function using the expectation value of measurements at the end of the circuit.
3. Optimization Method: Select and describe the optimization method used.
4. Classification Outcomes: Report the accuracy and other relevant metrics for train and test data.
5. Program Code: Provide the Python code used for the simulation.

Part 2: Extended Research (2/10 Credits)

Topics (choose one):

• Compare different classical optimization methods and/or loss functions.
• Justify the model using geometric representations like the Bloch sphere.
• Compare parameters, epochs, and structure with a classical neural network solving the same problem.
• Implement the parameter-shift rule for gradient evaluation.
• Apply the classifier to another binary classification problem with 3 inputs and report the outcomes.
• Any other interesting subject related to quantum machine learning (consultation with the instructor recommended).

Part 3: Extended Model for 5-input Parity Problem (2/10 Credits)

Requirements:

• Objective: Extend the quantum classifier model to handle the parity problem with 5 inputs.
• Data: Provided datasets (classA_train_N5.dat, classB_train_N5.dat, classA_test_N5.dat, classB_test_N5.dat).

Keywords

• Quantum Machine Learning
• Variational Quantum Circuit
• Quantum Classifier
• Binary Classification
• Parity Problem
• Quantum Optimization
• Quantum Cost Function
• Bloch Sphere
• Parameter-shift Rule
• Classical vs Quantum Models
• Quantum Data Preprocessing
• Quantum Simulation
• Python Quantum Programming
• Quantum Computing
• Quantum Neural Networks