Fleming-Viot helps speed up variational quantum algorithms in the presence of barren plateaus

Autor: Mastropietro, Daniel, Korpas, Georgios, Kungurtsev, Vyacheslav, Marecek, Jakub

Inspired by the Fleming-Viot stochastic process, we propose a parallel implementation of variational quantum algorithms with the aim of helping the algorithm get out of barren plateaus, where optimization direction is unclear. In the Fleming-Viot tra

Externí odkaz: http://arxiv.org/abs/2311.18090

A risk-based estimate of stable bands for frequency adaptive regularization in image restoration

Autor: Mastropietro, Daniel G.

Thesis (M.S.)--University of Wisconsin--Madison, 2000.
Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 74-75).

Externí odkaz: http://catalog.hathitrust.org/api/volumes/oclc/46591104.html

Boosting reinforcement learning with sparse and rare rewards using Fleming-Viot particle systems

Autor: Daniel Mastropietro, Szymon Majewski, Urtzi Ayesta, Matthieu Jonckheere

Publikováno v: HAL
à paraître
15th European Workshop on Reinforcement Learning (EWRL 2022)
15th European Workshop on Reinforcement Learning (EWRL 2022), Sep 2022, Milano, Italy

International audience; We consider reinforcement learning control problems under the average reward criterion in which non-zero rewards are both sparse and rare, that is, they occur in very few states and have a very small steady-state probability.

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::7b5c752d5597bb695bf794e40481a8a5
https://ut3-toulouseinp.hal.science/hal-03772025