18 July 2023 Tue 5 pm

Variational Quantum Algorithms (VQAs) are among the most promising candidates for achieving quantum advantage using noisy quantum devices. Their objective is to solve practical problems by utilizing parameterized quantum circuits in conjunction with classical optimization routines. However, VQAs can exhibit barren plateaus, where, under certain conditions, the gradient of the cost function vanishes exponentially with the number of qubits in the system, rendering the algorithm's training process infeasible. In this talk, we will discuss the fundamentals of VQAs and their key principles. We will then explore how the design of the ansatz, the selection of the cost function, and the impact of noise during the training process can influence the occurrence of barren plateaus. Additionally, we will show two distinct VQAs for determining the Geometric Measure of Entanglement (GME) in quantum states. These algorithms exhibit different behaviors regarding barren plateaus, highlighting how the same problem can become either easy or difficult to solve depending on the method employed.

< Previous

Next >