Reference : Kernel Regression over Graphs using Random Fourier Features
Scientific journals : Article
Engineering, computing & technology : Electrical & electronics engineering
Security, Reliability and Trust
http://hdl.handle.net/10993/50245
Kernel Regression over Graphs using Random Fourier Features
English
Elias, Vitor R.M. []
Gogineni, Vinay C. []
Alves Martins, Wallace mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SigCom >]
Werner, Stefan []
2022
IEEE Transactions on Signal Processing
Institute of Electrical and Electronics Engineers
Yes
International
1053-587X
United States
[en] kernel regression on graphs ; random Fourier features ; stochastic gradient ; recursive least squares
[en] This paper proposes efficient batch-based and online strategies for kernel regression over graphs (KRG). The proposed algorithms do not require the input signal to be a graph signal, whereas the target signal is defined over the graph. We first use random Fourier features (RFF) to tackle the complexity issues associated with kernel methods employed in the conventional KRG. For batch-based approaches, we also propose an implementation that reduces complexity by avoiding the inversion of large matrices. Then, we derive two distinct online strategies using RFF, namely, the mini-batch gradient KRG (MGKRG) and the recursive least squares KRG (RLSKRG). The stochastic gradient KRG (SGKRG) is introduced as a particular case of the MGKRG. The MGKRG and the SGKRG are low-complexity algorithms that employ stochastic gradient approximations in the regression-parameter update. The RLSKRG is a recursive implementation of the RFF-based batch KRG. A detailed stability analysis is provided for the proposed online algorithms, including convergence conditions in both mean and mean-squared senses. A discussion on complexity is also provided. Numerical simulations include a synthesized-data experiment and real-data experiments on temperature prediction, brain activity estimation, and image reconstruction. Results show that the RFF-based batch implementation offers competitive performance with a reduced computational burden when compared to the conventional KRG. The MGKRG offers a convenient trade-off between performance and complexity by varying the number of mini-batch samples. The RLSKRG has a faster convergence than the MGKRG and matches the performance of the batch implementation.
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Grant number: 88887.310189/2018-00 ; FAPERJ - Brasil ; Research Council of Norway
Researchers ; Professionals ; Students
http://hdl.handle.net/10993/50245
10.1109/TSP.2022.3149134
H2020 ; 742648 - AGNOSTIC - Actively Enhanced Cognition based Framework for Design of Complex Systems

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Limited access
IEEE-TSP-KRG-2022.pdfAuthor postprint1.9 MBRequest a copy

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.