Signal reconstruction by means of Embedding, Clustering and AutoEncoder Ensembles

Corrado Mio, Gabriele Gianini

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

We study the denoising and reconstruction of corrupted signals by means of AutoEncoder ensembles. In order to guarantee experts' diversity in the ensemble, we apply, prior to learning, a dimensional reduction pass (to map the examples into a suitable Euclidean space) and a partitional clustering pass: each cluster is then used to train a distinct AutoEncoder. We study the approach with an audio file benchmark: the original signals are artificially corrupted by Doppler effect and reverb. The results support the comparative effectiveness of the approach, w.r.t. the approach based on a single AutoEncoder. The processing pipeline using Local Linear Embedding, k means, then k Convolutional Denoising AutoEncoders reduces the reconstruction error by 35% w.r.t. the baseline approach.

Original languageBritish English
Title of host publicationProceedings - IEEE Symposium on Computers and Communications
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728129990
DOIs
StatePublished - Jun 2019
Event2019 IEEE Symposium on Computers and Communications, ISCC 2019 - Barcelona, Spain
Duration: 29 Jun 20193 Jul 2019

Publication series

NameProceedings - IEEE Symposium on Computers and Communications
Volume2019-June
ISSN (Print)1530-1346

Conference

Conference2019 IEEE Symposium on Computers and Communications, ISCC 2019
Country/TerritorySpain
CityBarcelona
Period29/06/193/07/19

Fingerprint

Dive into the research topics of 'Signal reconstruction by means of Embedding, Clustering and AutoEncoder Ensembles'. Together they form a unique fingerprint.

Cite this