Pruning Latent Neurons in Autoencoders using Early-Bird Tickets

Master Thesis (2022)
Author(s)

R.F. Klazinga (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Contributor(s)

Marco Loog – Mentor (TU Delft - Pattern Recognition and Bioinformatics)

J.C. van Gemert – Graduation committee member (TU Delft - Pattern Recognition and Bioinformatics)

CCS Liem – Graduation committee member (TU Delft - Multimedia Computing)

Faculty
Electrical Engineering, Mathematics and Computer Science
Copyright
© 2022 Rembrandt Klazinga
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 Rembrandt Klazinga
Graduation Date
10-06-2022
Awarding Institution
Delft University of Technology
Programme
Computer Science
Faculty
Electrical Engineering, Mathematics and Computer Science
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Autoencoders seek to encode their input into a bottleneck of latent neurons, and then decode it to reconstruct the input. However, if the input data has an intrinsic dimension (ID) smaller than the number of latent neurons in the bottleneck, this encoding becomes redundant.
In this paper, we study using the Early-Bird (EB) technique, a structural pruning method, to regularise and prune the redundant latent neurons. We do this for both linear-layer and convolutional autoencoders, on 1D and 2D data. We find that increasing the strength of EB regularisation specifically on the latent layer can lead to all redundant latent neurons (and no more) being removed in one training run.
We also compare using EB in this manner to existing ID estimation methods: we find it performs comparable to older methods like local-PCA, also being relatively robust to noise, but that it does not match the best existing ID estimation methods.

Files

Thesis_RKlazinga.pdf
(pdf | 1.58 Mb)
License info not available