Quantum Computing + AI: Advanced
Variational algorithms, quantum machine learning, real hardware and error correction — in about six hours
For learners who have completed the Foundation course (or pass the placement test) and want hands-on depth: tunable quantum circuits, quantum machine learning, making noisy hardware useful, and running real circuits on IBM quantum computers. Same short-lesson format as Foundation — a video, a short read and a knowledge check per lesson — just one level up.
What you'll be able to do
- ✓Describe mixed states and noise with density matrices, and simulate simple noise yourself
- ✓Build and train a parameterised (variational) circuit, and explain VQE and QAOA
- ✓Explain why some variational circuits are hard to train (barren plateaus)
- ✓Build quantum kernels and quantum neural networks, and evaluate them against classical baselines
- ✓Apply basic error-mitigation techniques and explain how error correction protects a logical qubit
- ✓Explain Shor's algorithm end to end and run a circuit on real IBM quantum hardware
- ✓Critically evaluate a quantum-advantage or QML claim using a five-point checklist
Syllabus
Module 1
Foundation Refresher
A fast recap of Foundation, plus exactly the extra math and notation Advanced needs.
⌄
Module 1
Foundation Refresher
A fast recap of Foundation, plus exactly the extra math and notation Advanced needs.
You will learn to
- Restate superposition, interference, entanglement and circuits in one place
- Work with complex numbers, inner products and unitary matrices
- Read Dirac notation and distinguish a superposition from a mixed state
- ·Quick recap: superposition, gates and entanglement10 min
- ·Complex numbers, vectors and matrices15 min
- ·Dirac notation and a first look at mixed states14 min
Module 2
Density Matrices and Noise
The tool for describing imperfect qubits, and your first hands-on noise simulation.
⌄
Module 2
Density Matrices and Noise
The tool for describing imperfect qubits, and your first hands-on noise simulation.
You will learn to
- Represent a mixed state with a density matrix
- Describe noise as a quantum channel, informally
- Simulate simple readout noise and see its effect on measurement counts
- ·Mixed states and the density matrix15 min
- ·Quantum channels: modelling noise15 min
- ·Fidelity and simulating noise16 min
Module 3
Variational Algorithms: VQE and QAOA
Tunable circuits trained like a machine-learning model, applied to chemistry and optimisation.
⌄
Module 3
Variational Algorithms: VQE and QAOA
Tunable circuits trained like a machine-learning model, applied to chemistry and optimisation.
You will learn to
- Build a parameterised circuit and explain the variational principle
- Explain how VQE estimates a molecule's ground-state energy
- Encode a Max-Cut problem as a cost function and solve it with QAOA
- ·The variational principle and parameterised circuits15 min
- ·VQE: estimating a molecule's energy18 min
- ·QAOA: solving Max-Cut17 min
Module 4
Training Parameterised Circuits
How gradients work on quantum hardware, and why bigger circuits get harder to train.
⌄
Module 4
Training Parameterised Circuits
How gradients work on quantum hardware, and why bigger circuits get harder to train.
You will learn to
- Explain the parameter-shift rule for computing a quantum gradient
- Recognise a barren plateau and why it makes training stall
- ·The parameter-shift rule15 min
- ·Barren plateaus15 min
Module 5
Quantum Machine Learning: Kernels and Networks
Two ways to build a QML model, and what to check before trusting either one.
⌄
Module 5
Quantum Machine Learning: Kernels and Networks
Two ways to build a QML model, and what to check before trusting either one.
You will learn to
- Build a quantum feature map and use it as a kernel
- Describe a quantum neural network and how it trains inside a classical framework
- ·Quantum feature maps and kernels15 min
- ·Quantum neural networks and hybrid models15 min
Module 6
Making Noisy Hardware Useful
Error mitigation you can use today, and the error-correction ideas behind tomorrow's machines.
⌄
Module 6
Making Noisy Hardware Useful
Error mitigation you can use today, and the error-correction ideas behind tomorrow's machines.
You will learn to
- Apply basic error-mitigation techniques and estimate a noiseless result
- Explain how a repetition code detects an error without measuring the data
- Describe the surface code and what "below threshold" means
- ·Error mitigation in practice16 min
- ·Error correction basics15 min
- ·The surface code and thresholds15 min
Module 7
Shor's Algorithm and Real Hardware
Factoring end to end, its consequences for cryptography, and running your own circuit on an IBM quantum computer.
⌄
Module 7
Shor's Algorithm and Real Hardware
Factoring end to end, its consequences for cryptography, and running your own circuit on an IBM quantum computer.
You will learn to
- Trace Shor's algorithm from factoring to period-finding to the answer
- Explain why post-quantum cryptography migration is happening now, not later
- Submit a circuit to real IBM quantum hardware and interpret the result
- ·Shor's algorithm, end to end18 min
- ·Post-quantum cryptography migration13 min
- ·Running on real IBM quantum hardware20 min
Module 8
AI for Quantum, and Critical Reading
Where AI is already helping quantum computers, and a checklist for reading any advantage claim.
⌄
Module 8
AI for Quantum, and Critical Reading
Where AI is already helping quantum computers, and a checklist for reading any advantage claim.
You will learn to
- Name concrete ways machine learning already helps quantum hardware
- Apply a five-point checklist to a quantum-advantage or QML claim
- ·AI for quantum: decoders, calibration and compilation14 min
- ·Dequantisation and reading a QML paper critically15 min
Capstone
Pick one: run VQE on a toy two-level system or QAOA on a small Max-Cut instance in the simulator, then submit the same circuit to real IBM quantum hardware. Write a short report comparing the ideal, simulated and hardware results, and explain the gap in terms of noise.