Trainability of Dissipative Perceptron-Based Quantum Neural Networks

Several architectures have been proposed for quantum neural networks (QNNs), with the goal of efficiently performing machine learning tasks on quantum data. Rigorous scaling results are urgently needed for specific QNN constructions to understand which, if any, will be trainable at a large scale. He...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Physical review letters Jg. 128; H. 18; S. 180505
Hauptverfasser: Sharma, Kunal, Cerezo, M., Cincio, Lukasz, Coles, Patrick J.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States American Physical Society (APS) 06.05.2022
Schlagworte:
artificial neural networks
CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS
machine learning
quantum algorithms
quantum computation
quantum information processing
quantum networks
ISSN:0031-9007, 1079-7114, 1079-7114
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Several architectures have been proposed for quantum neural networks (QNNs), with the goal of efficiently performing machine learning tasks on quantum data. Rigorous scaling results are urgently needed for specific QNN constructions to understand which, if any, will be trainable at a large scale. Here, we analyze the gradient scaling (and hence the trainability) for a recently proposed architecture that we call dissipative QNNs (DQNNs), where the input qubits of each layer are discarded at the layer's output. We find that DQNNs can exhibit barren plateaus, i.e., gradients that vanish exponentially in the number of qubits. Moreover, we provide quantitative bounds on the scaling of the gradient for DQNNs under different conditions, such as different cost functions and circuit depths, and show that trainability is not always guaranteed. Our work represents the first rigorous analysis of the scalability of a perceptron-based QNN.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
89233218CNA000001
LA-UR-20-23484
USDOE Laboratory Directed Research and Development (LDRD) Program
ISSN:0031-9007
1079-7114
1079-7114
DOI:10.1103/PhysRevLett.128.180505