File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1016/j.neunet.2023.01.042
- Scopus: eid_2-s2.0-85147416242
- PMID: 36745940
- WOS: WOS:000935345900001
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: A feedforward unitary equivariant neural network
Title | A feedforward unitary equivariant neural network |
---|---|
Authors | |
Keywords | Equivariant neural network Feedforward neural network Rotational equivariant Unitary equivariant |
Issue Date | 4-Feb-2023 |
Publisher | Elsevier |
Citation | Neural Networks, 2023, v. 161, p. 154-164 How to Cite? |
Abstract | We devise a new type of feedforward neural network. It is equivariant with respect to the unitary group U(n). The input and output can be vectors in Cn with arbitrary dimension n. No convolution layer is required in our implementation. We avoid errors due to truncated higher order terms in Fourier-like transformation. The implementation of each layer can be done efficiently using simple calculations. As a proof of concept, we have given empirical results on the prediction of the dynamics of atomic motion to demonstrate the practicality of our approach.(c) 2023 Published by Elsevier Ltd. |
Persistent Identifier | http://hdl.handle.net/10722/338114 |
ISSN | 2023 Impact Factor: 6.0 2023 SCImago Journal Rankings: 2.605 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Ma, PW | - |
dc.contributor.author | Chan, THH | - |
dc.date.accessioned | 2024-03-11T10:26:22Z | - |
dc.date.available | 2024-03-11T10:26:22Z | - |
dc.date.issued | 2023-02-04 | - |
dc.identifier.citation | Neural Networks, 2023, v. 161, p. 154-164 | - |
dc.identifier.issn | 0893-6080 | - |
dc.identifier.uri | http://hdl.handle.net/10722/338114 | - |
dc.description.abstract | We devise a new type of feedforward neural network. It is equivariant with respect to the unitary group U(n). The input and output can be vectors in Cn with arbitrary dimension n. No convolution layer is required in our implementation. We avoid errors due to truncated higher order terms in Fourier-like transformation. The implementation of each layer can be done efficiently using simple calculations. As a proof of concept, we have given empirical results on the prediction of the dynamics of atomic motion to demonstrate the practicality of our approach.(c) 2023 Published by Elsevier Ltd. | - |
dc.language | eng | - |
dc.publisher | Elsevier | - |
dc.relation.ispartof | Neural Networks | - |
dc.rights | This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. | - |
dc.subject | Equivariant neural network | - |
dc.subject | Feedforward neural network | - |
dc.subject | Rotational equivariant | - |
dc.subject | Unitary equivariant | - |
dc.title | A feedforward unitary equivariant neural network | - |
dc.type | Article | - |
dc.identifier.doi | 10.1016/j.neunet.2023.01.042 | - |
dc.identifier.pmid | 36745940 | - |
dc.identifier.scopus | eid_2-s2.0-85147416242 | - |
dc.identifier.volume | 161 | - |
dc.identifier.spage | 154 | - |
dc.identifier.epage | 164 | - |
dc.identifier.eissn | 1879-2782 | - |
dc.identifier.isi | WOS:000935345900001 | - |
dc.publisher.place | OXFORD | - |
dc.identifier.issnl | 0893-6080 | - |