Neural network tuning
This video explains the concept of tuning with neural networks and outlines basic machine learning tuning workflow.
Prerequisite knowledge
Basic knowledge of Neural Networks. The Machine Learning Primer from Module 1 of the course Machine Learning for Semiconductor Quantum Devices is sufficient.
Main takeaways
- There is fundamental difference between fixed algorithms and learning from data.
- Fast evaluation and good generalization properties are key advantages of machine learning algorithms used in experimental workflows.
Further thinking
True or False: Machine learning algorithm can in principle take any type of measurement data as an input.
Further reading
Very nice general overview of quantum dot tuning by Justyna Zwolak and Jake Taylor https://journals.aps.org/rmp/abstract/10.1103/RevModPhys.95.011006