Hi all,
Tomorrow Moritz Scheer will tell us about his master thesis entitled "Quantum Advantage in Derivative Pricing & Variationally Trained Gaussian Loaders". See below for the abstract. The talk will take place at 2pm in HIT H42.
Best, Ladina
*** Title: Quantum Advantage in Derivative Pricing & Variationally Trained Gaussian Loaders
Abstract: Loading of probability distributions is an important subroutine in many quantum algorithms, necessary for example for pricing derivatives in finance. This talk gives an introduction to derivative pricing and explains its working-principle using European call options and auto-callable options as examples. It shows how a quadratic quantum speedup can be achieved by accelerating classical Monte-Carlo methods using amplitude estimation. An important subroutine in derivative pricing is the loading of Gaussian probability distributions into the quantum computer. The second part of the talk introduces variationally trained Gaussian loaders and presents a successful implementation both with the real-amplitudes ansatz and the variational Hamiltonian ansatz under ideal conditions. It is shown that this implementation struggles when taking the effects of estimation noise into account.