PMID- 26654210
OWN - NLM
STAT- MEDLINE
DCOM- 20161014
LR  - 20191210
IS  - 1530-888X (Electronic)
IS  - 0899-7667 (Linking)
VI  - 28
IP  - 2
DP  - 2016 Feb
TI  - Correlational Neural Networks.
PG  - 257-85
LID - 10.1162/NECO_a_00801 [doi]
AB  - Common representation learning (CRL), wherein different descriptions (or views) 
      of the data are embedded in a common subspace, has been receiving a lot of 
      attention recently. Two popular paradigms here are canonical correlation analysis 
      (CCA)-based approaches and autoencoder (AE)-based approaches. CCA-based 
      approaches learn a joint representation by maximizing correlation of the views 
      when projected to the common subspace. AE-based methods learn a common 
      representation by minimizing the error of reconstructing the two views. Each of 
      these approaches has its own advantages and disadvantages. For example, while 
      CCA-based approaches outperform AE-based approaches for the task of transfer 
      learning, they are not as scalable as the latter. In this work, we propose an 
      AE-based approach, correlational neural network (CorrNet), that explicitly 
      maximizes correlation among the views when projected to the common subspace. 
      Through a series of experiments, we demonstrate that the proposed CorrNet is 
      better than AE and CCA with respect to its ability to learn correlated common 
      representations. We employ CorrNet for several cross-language tasks and show that 
      the representations learned using it perform better than the ones learned using 
      other state-of-the-art approaches.
FAU - Chandar, Sarath
AU  - Chandar S
AD  - University of Montreal, Montreal QC H3T 1J4, Canada apsarathchandar@gmail.com.
FAU - Khapra, Mitesh M
AU  - Khapra MM
AD  - IBM Research India, Bangalore 560077, India mikhapra@in.ibm.com.
FAU - Larochelle, Hugo
AU  - Larochelle H
AD  - University of Sherbrooke, Sherbrooke QC J1K 2R1, Canada 
      hugo.larochelle@usherbrooke.ca.
FAU - Ravindran, Balaraman
AU  - Ravindran B
AD  - Indian Institute of Technology Madras, Chennai, India, 600036 
      ravi@cse.iitm.ac.in.
LA  - eng
PT  - Journal Article
DEP - 20151214
PL  - United States
TA  - Neural Comput
JT  - Neural computation
JID - 9426182
SB  - IM
MH  - Humans
MH  - *Learning
MH  - Models, Neurological
MH  - *Neural Networks, Computer
MH  - Pattern Recognition, Automated/*methods
MH  - *Statistics as Topic
EDAT- 2015/12/15 06:00
MHDA- 2016/10/16 06:00
CRDT- 2015/12/15 06:00
PHST- 2015/12/15 06:00 [entrez]
PHST- 2015/12/15 06:00 [pubmed]
PHST- 2016/10/16 06:00 [medline]
AID - 10.1162/NECO_a_00801 [doi]
PST - ppublish
SO  - Neural Comput. 2016 Feb;28(2):257-85. doi: 10.1162/NECO_a_00801. Epub 2015 Dec 
      14.