Multi-Armed Bandits and Quantum Channel Oracles

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Buchholz, Simon, Kübler, Jonas M., Schölkopf, Bernhard
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909550389493760
author Buchholz, Simon
Kübler, Jonas M.
Schölkopf, Bernhard
author_facet Buchholz, Simon
Kübler, Jonas M.
Schölkopf, Bernhard
contents Multi-armed bandits are one of the theoretical pillars of reinforcement learning. Recently, the investigation of quantum algorithms for multi-armed bandit problems was started, and it was found that a quadratic speed-up (in query complexity) is possible when the arms and the randomness of the rewards of the arms can be queried in superposition. Here we introduce further bandit models where we only have limited access to the randomness of the rewards, but we can still query the arms in superposition. We show that then the query complexity is the same as for classical algorithms. This generalizes the prior result that no speed-up is possible for unstructured search when the oracle has positive failure probability.
format Preprint
id arxiv_https___arxiv_org_abs_2301_08544
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Multi-Armed Bandits and Quantum Channel Oracles
Buchholz, Simon
Kübler, Jonas M.
Schölkopf, Bernhard
Quantum Physics
Machine Learning
68Q12
Multi-armed bandits are one of the theoretical pillars of reinforcement learning. Recently, the investigation of quantum algorithms for multi-armed bandit problems was started, and it was found that a quadratic speed-up (in query complexity) is possible when the arms and the randomness of the rewards of the arms can be queried in superposition. Here we introduce further bandit models where we only have limited access to the randomness of the rewards, but we can still query the arms in superposition. We show that then the query complexity is the same as for classical algorithms. This generalizes the prior result that no speed-up is possible for unstructured search when the oracle has positive failure probability.
title Multi-Armed Bandits and Quantum Channel Oracles
topic Quantum Physics
Machine Learning
68Q12
url https://arxiv.org/abs/2301.08544