TY - JOUR
T1 - Semisupervised Affinity Matrix Learning via Dual-Channel Information Recovery
AU - Jia, Yuheng
AU - Liu, Hui
AU - Hou, Junhui
AU - Kwong, Sam
AU - Zhang, Qingfu
PY - 2022/8
Y1 - 2022/8
N2 - This article explores the problem of semisupervised
affinity matrix learning, that is, learning an affinity matrix of
data samples under the supervision of a small number of pairwise
constraints (PCs). By observing that both the matrix encoding
PCs, called pairwise constraint matrix (PCM) and the empirically constructed affinity matrix (EAM), express the similarity
between samples, we assume that both of them are generated
from a latent affinity matrix (LAM) that can depict the ideal
pairwise relation between samples. Specifically, the PCM can be
thought of as a partial observation of the LAM, while the EAM
is a fully observed one but corrupted with noise/outliers. To this
end, we innovatively cast the semisupervised affinity matrix learning as the recovery of the LAM guided by the PCM and EAM,
which is technically formulated as a convex optimization problem.
We also provide an efficient algorithm for solving the resulting model numerically. Extensive experiments on benchmark
datasets demonstrate the significant superiority of our method
over state-of-the-art ones when used for constrained clustering
and dimensionality reduction. The code is publicly available at
https://github.com/jyh-learning/LAM.
AB - This article explores the problem of semisupervised
affinity matrix learning, that is, learning an affinity matrix of
data samples under the supervision of a small number of pairwise
constraints (PCs). By observing that both the matrix encoding
PCs, called pairwise constraint matrix (PCM) and the empirically constructed affinity matrix (EAM), express the similarity
between samples, we assume that both of them are generated
from a latent affinity matrix (LAM) that can depict the ideal
pairwise relation between samples. Specifically, the PCM can be
thought of as a partial observation of the LAM, while the EAM
is a fully observed one but corrupted with noise/outliers. To this
end, we innovatively cast the semisupervised affinity matrix learning as the recovery of the LAM guided by the PCM and EAM,
which is technically formulated as a convex optimization problem.
We also provide an efficient algorithm for solving the resulting model numerically. Extensive experiments on benchmark
datasets demonstrate the significant superiority of our method
over state-of-the-art ones when used for constrained clustering
and dimensionality reduction. The code is publicly available at
https://github.com/jyh-learning/LAM.
KW - Clustering
KW - graph learning
KW - semisupervised learning
UR - http://www.scopus.com/inward/record.url?scp=85099547332&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85099547332&origin=recordpage
U2 - 10.1109/TCYB.2020.3041493
DO - 10.1109/TCYB.2020.3041493
M3 - 21_Publication in refereed journal
VL - 52
SP - 7919
EP - 7930
JO - IEEE Transactions on Cybernetics
JF - IEEE Transactions on Cybernetics
SN - 2168-2267
IS - 8
ER -