Quantum Physics
[Submitted on 30 Sep 2020 (this version), latest version 19 Oct 2021 (v3)]
Title:Barren plateaus preclude learning scramblers
View PDFAbstract:Scrambling, the rapid spread of entanglement and information through many-body quantum systems, is of fundamental importance to thermalization and quantum chaos but can be challenging to investigate using standard techniques. Recently, quantum machine learning (QML) has emerged as a promising platform to study complex quantum processes. This prompts the question of whether QML could be used to investigate scrambling. In this Letter, we prove a no-go theorem constraining the use of QML to learn an unknown scrambling process. Specifically, we show that employing QML to learn an unknown scrambling unitary, for any choice of variational ansatz, will with high probability lead to a barren plateau landscape, i.e., where the cost gradient vanishes exponentially in the system size. We provide numerics extending our results to approximate scramblers. More broadly, our no-go theorem implies that prior information is required to learn an unknown unitary and thus places limits on the learnability of quantum processes.
Submission history
From: Zoe Holmes [view email][v1] Wed, 30 Sep 2020 17:31:36 UTC (784 KB)
[v2] Wed, 12 May 2021 20:45:04 UTC (1,127 KB)
[v3] Tue, 19 Oct 2021 04:17:10 UTC (1,127 KB)
Current browse context:
quant-ph
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.