Close

Filter courses

Show courses Show filters
Close

Germanium Quantum Technology

01 /What can CMOS technology do for quantum computing? 02 /Semiconductor devices and scaling with industrial approach 03 /Semiconductor foundry facilities: the good and the bad for qubits 04 /IMEC's latest qubit developments 05 /Summary and outlook 06 /Semiconductors for spin qubits 07 /Germanium quantum wells on silicon 08 /Germanium quantum wells on silicon-germanium 09 /Growth methods 10 /Characterization techniques 11 /Electrical characterization of germanium quantum wells 12 /Fabrication of H-FET 13 /Semiconductor qubits: vision and challenges 14 /Quantum materials 15 /Quantum dot qubits and their operation 16 /Types of quantum dot qubits 17 /Qubit Readout 18 /Qubit Control 19 /Qubit Coherence and Fidelity 20 /Multi-qubit control and quantum algorithms 21 /Quantum links and scaling 22 /Physics of holes 23 /Hole spin qubits 24 /Introduction to electronics for quantum computing 25 /Room temperature electronics 26 /Cryogenic qubit chip carriers 27 /Contribution to IGNITE 28 /Quantum Control Stack for Spin Qubits 29 /Auto-tuning a quantum computer 30 /Tuning and operation of arrays 31 /Finding operation points example 32 /What is auto-tuning? 33 /Neural network tuning 34 /Navigating charge stability diagrams 35 /Experimental implementation 36 /Towards universal quantum algorithms 37 /Classical error correction 38 /Quantum error correction 39 /Progress and challenges 40 /Introduction to quantum algorithms 41 /The first algorithms 42 /Quantum annealing 43 /Quantum Machine Learning

Auto-tuning a quantum computer

For quantum-dot-based qubits, while the controllability over the electric potential given by the gate electrodes is immense, each electrode presents a degree of freedom that must be tuned to a value that traps the desired number of electrons or holes.

In this video, we elucidate this roadblock and discuss a potential solution: instead of tuning all these electrodes by hand, which rapidly becomes difficult, we could use automated techniques. In particular, by cleverly framing the problem, we could use machine learning methods to achieve automated tuning.

First, we discuss why we need gate voltages in the first place and how these gate electrodes actually trap our qubit arrays.

Prerequisite knowledge

  • Spin qubits
  • Qubit decoherence
  • Quantum dot qubits
  • Operations on spin qubits
  • Capturing a Single Electron

Main takeaways

Tuning is becoming a significant roadblock for quantum-dot-based quantum processors and needs a lot of human attention.

Automated methods, whether optimization algorithms or machine learning, can be used to tune these gate electrodes.

Framing the problem in a clever way is essential for choosing the right algorithm that is scalable, rather than brute-force methods.