Show simple item record

dc.contributor.authorNichele, Stefano
dc.contributor.authorMolund, Andreas
dc.date.accessioned2018-05-09T07:58:01Z
dc.date.accessioned2018-09-14T07:24:56Z
dc.date.available2018-05-09T07:58:01Z
dc.date.available2018-09-14T07:24:56Z
dc.date.issued2017
dc.identifier.citationNichele S, Molund. Deep learning with cellular automaton-based reservoir computing. Complex Systems. 2017;26(4):319-340en
dc.identifier.issn0891-2513
dc.identifier.urihttps://hdl.handle.net/10642/6164
dc.description.abstractRecurrent neural networks (RNNs) have been a prominent concept wiithin artificial intelligence. They are inspired by biological neural net works (BNNs) and provide an intuitive and abstract representation of how BNNs work. Derived from the more generic artificial neural networks (ANNs), the recurrent ones are meant to be used for temporal tasks, such as speech recognition, because they are capable of memorizing historic input. However, such networks are very time consuming to train as a result of their inherent nature. Recently, echo state Networks and liquid state machines have been proposed as possible RNN alternatives, under the name of reservoir computing (RC). Reservoir computers are far easier to train. In this paper, cellular automata (CAs) are used as a reservoir and are tested on the five-bit memory task (a well known benchmark within the RC community). The work herein provides a method of mapping binary inputs from the task onto the automata and a recurrent architecture for handling the sequential aspects. Furthermore, a layered (deep) reservoir architecture is proposed. Performances are compared to earlier work, in addition to the single-layer version. Results show that the single cellular automaton (CA) reservoir system yields similar results to state-of-the-art work. The system comprised of two layered reservoirs does show a noticeable improvement compared to a single CA reservoir. This work lays the foundation for implementations of deep learning with CA-based reservoir systems.en
dc.language.isoenen
dc.publisherComplex Systems Publications Incen
dc.relation.ispartofseriesComplex Systems;Volume 26, Issue 4
dc.relation.urihttps://arxiv.org/pdf/1703.02806.pdf
dc.rightsen
dc.subjectDeep learningen
dc.subjectCellular automatonsen
dc.subjectReservoir computingen
dc.titleDeep learning with cellular automaton-based reservoir computingen
dc.typeJournal articleen
dc.typePeer revieweden
dc.date.updated2018-05-09T07:58:01Z
dc.description.versionpublishedVersionen
dc.identifier.doihttp://dx.doi.org/10.25088/ComplexSystems.26.4.319
dc.identifier.cristinID1525801
dc.source.issn0891-2513
dc.relation.journalComplex Systems


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record