Back to Search
Start Over
Scalable out-of-sample extension of graph embeddings using deep neural networks
- Source :
- Pattern Recognition Letters. 94:1-6
- Publication Year :
- 2017
- Publisher :
- Elsevier BV, 2017.
-
Abstract
- Several popular graph embedding techniques for representation learning and dimensionality reduction rely on performing computationally expensive eigendecompositions to derive a nonlinear transformation of the input data space. The resulting eigenvectors encode the embedding coordinates for the training samples only, and so the embedding of novel data samples requires further costly computation. In this paper, we present a method for the out-of-sample extension of graph embeddings using deep neural networks (DNN) to parametrically approximate these nonlinear maps. Compared with traditional nonparametric out-of-sample extension methods, we demonstrate that the DNNs can generalize with equal or better fidelity and require orders of magnitude less computation at test time. Moreover, we find that unsupervised pretraining of the DNNs improves optimization for larger network sizes, thus removing sensitivity to model selection.<br />Comment: 10 pages, 2 figures, 1 table, this paper is under consideration for publication in Pattern Recognition Letters
- Subjects :
- FOS: Computer and information sciences
Graph embedding
Computer science
Machine Learning (stat.ML)
02 engineering and technology
Machine Learning (cs.LG)
Methodology (stat.ME)
030507 speech-language pathology & audiology
03 medical and health sciences
Statistics - Machine Learning
Artificial Intelligence
0202 electrical engineering, electronic engineering, information engineering
Neural and Evolutionary Computing (cs.NE)
Statistics - Methodology
Model selection
Dimensionality reduction
Computer Science - Neural and Evolutionary Computing
Graph
Manifold
Computer Science - Learning
Signal Processing
Scalability
Graph (abstract data type)
Embedding
020201 artificial intelligence & image processing
Computer Vision and Pattern Recognition
0305 other medical science
Algorithm
Feature learning
Software
Subjects
Details
- ISSN :
- 01678655
- Volume :
- 94
- Database :
- OpenAIRE
- Journal :
- Pattern Recognition Letters
- Accession number :
- edsair.doi.dedup.....72b8a1a3808adddb56546887028d881e