Autoencoder node saliency: Selecting relevant latent representations
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
The autoencoder is an artificial neural network that performs nonlinear dimension reduction and learns hidden representations of unlabeled data. With a linear transfer function it is similar to the principal component analysis (PCA). While both methods use weight vectors for linear transformations, the autoencoder does not come with any indication similar to the eigenvalues in PCA that are paired with eigenvectors. In this work, we propose a novel autoencoder node saliency method that examines whether the features constructed by autoencoders exhibit properties related to known class labels. The supervised node saliency ranks the nodes based on their capability of performing a learning task. It is coupled with the normalized entropy difference (NED). We establish a property for NED values to verify classifying behaviors among the top ranked nodes. Lastly, by applying our methods to real datasets, we demonstrate their ability to provide indications on the performing nodes and explain the learned tasks in autoencoders.
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- Grant/Contract Number:
- AC52-07NA27344
- OSTI ID:
- 1491664
- Alternate ID(s):
- OSTI ID: 1636754
- Report Number(s):
- LLNL-JRNL--741590; 896098
- Journal Information:
- Pattern Recognition, Journal Name: Pattern Recognition Journal Issue: C Vol. 88; ISSN 0031-3203
- Publisher:
- ElsevierCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Assessment of Autoencoder Architectures for Data Representation
|
book | October 2019 |
Discriminative stacked autoencoder for feature representation and classification
|
journal | January 2020 |
A Robust Deep Learning Approach for Spatiotemporal Estimation of Satellite AOD and PM2.5
|
journal | January 2020 |
Similar Records
Latent Representation Learning for Structural Characterization of Catalysts