PMID- 33746467
OWN - NLM
STAT- PubMed-not-MEDLINE
LR  - 20210324
IS  - 1053-587X (Print)
IS  - 1053-587X (Linking)
VI  - 68
DP  - 2020
TI  - Graph-based Learning under Perturbations via Total Least-Squares.
PG  - 2870-2882
LID - 10.1109/tsp.2020.2982833 [doi]
AB  - Graphs are pervasive in different fields unveiling complex relationships between 
      data. Two major graph-based learning tasks are topology identification and 
      inference of signals over graphs. Among the possible models to explain data 
      interdependencies, structural equation models (SEMs) accommodate a gamut of 
      applications involving topology identification. Obtaining conventional SEMs 
      though requires measurements across nodes. On the other hand, typical signal 
      inference approaches 'blindly trust' a given nominal topology. In practice 
      however, signal or topology perturbations may be present in both tasks, due to 
      model mismatch, outliers, outages or adversarial behavior. To cope with such 
      perturbations, this work introduces a regularized total least-squares (TLS) 
      approach and iterative algorithms with convergence guarantees to solve both 
      tasks. Further generalizations are also considered relying on structured and/or 
      weighted TLS when extra prior information on the perturbation is available. 
      Analyses with simulated and real data corroborate the effectiveness of the novel 
      TLS-based approaches.
FAU - Ceci, Elena
AU  - Ceci E
AD  - Department of Information Engineering, Electronics and Telecommunications, 
      Sapienza University of Rome, Rome 00184, Italy.
FAU - Shen, Yanning
AU  - Shen Y
AD  - CPCC and the Department of Electrical Engineering and Computer Science, the 
      University of California, Irvine, 92697, USA.
FAU - Giannakis, Georgios B
AU  - Giannakis GB
AD  - Digital Technology Center and the Department of Electrical and Computer 
      Engineering, University of Minnesota, Minneapolis, MN 55455 USA.
FAU - Barbarossa, Sergio
AU  - Barbarossa S
AD  - Department of Information Engineering, Electronics and Telecommunications, 
      Sapienza University of Rome, Rome 00184, Italy.
LA  - eng
GR  - R01 GM104975/GM/NIGMS NIH HHS/United States
PT  - Journal Article
DEP - 20200323
PL  - United States
TA  - IEEE Trans Signal Process
JT  - IEEE transactions on signal processing : a publication of the IEEE Signal 
      Processing Society
JID - 9885223
PMC - PMC7971163
MID - NIHMS1598219
OTO - NOTNLM
OT  - Graph and signal perturbations
OT  - graph signal reconstruction
OT  - structural equation models
OT  - topology identification
OT  - total least-squares
EDAT- 2020/01/01 00:00
MHDA- 2020/01/01 00:01
PMCR- 2021/03/18
CRDT- 2021/03/22 06:57
PHST- 2021/03/22 06:57 [entrez]
PHST- 2020/01/01 00:00 [pubmed]
PHST- 2020/01/01 00:01 [medline]
PHST- 2021/03/18 00:00 [pmc-release]
AID - 10.1109/tsp.2020.2982833 [doi]
PST - ppublish
SO  - IEEE Trans Signal Process. 2020;68:2870-2882. doi: 10.1109/tsp.2020.2982833. Epub 
      2020 Mar 23.