Skip to content

Latest commit

 

History

History
7 lines (4 loc) · 1.1 KB

14 How to: QML Loss, Optimizer.md

File metadata and controls

7 lines (4 loc) · 1.1 KB

How to Select the Correct QML Loss Function and Optimizer PDF + Discussion 10/12/23.

Quantum machine learning uses loss functions and optimizer algorithms similar to classical deep learning methods. Since quantum circuits are treated as Python functions that return classical expectation values with the help of Qiskit or PennyLane libraries; hybrid model parameters are optimized in ways understandable to data scientists.

Here, the importance of gradient descent and adaptive methods are detailed in regards to navigating loss landscapes. Examples from Manning, DeepLearning AI, and ChemicalQDevice highlight differences in classical or hybrid model approaches. Recommendations to select correct loss functions and optimizers are provided for major platforms; which are available through PyTorch, Keras, or NumPy.

Discussion, GitHub study.