Skip to main navigation Skip to search Skip to main content

Universality of Real Minimal Complexity Reservoir

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Reservoir Computing (RC) models, a subclass of recurrent neural networks, are distinguished by their fixed, non-trainable input layer and dynamically coupled reservoir, with only the static readout layer being trained. This design circumvents the issues associated with backpropagating error signals through time, thereby enhancing both stability and training efficiency. RC models have been successfully applied across a broad range of application domains. Crucially, they have been demonstrated to be universal approximators of time-invariant dynamic filters with fading memory, under various settings of approximation norms and input driving sources. Simple Cycle Reservoirs (SCR) represent a specialized class of RC models with a highly constrained reservoir architecture, characterized by uniform ring connectivity and binary input-to-reservoir weights with an aperiodic sign pattern. For linear reservoirs, given the reservoir size, the reservoir construction has only one degree of freedom -- the reservoir cycle weight. Such architectures are particularly amenable to hardware implementations without significant performance degradation in many practical tasks. In this study we endow these observations with solid theoretical foundations by proving that SCRs operating in real domain are universal approximators of time-invariant dynamic filters with fading memory. Our results supplement recent research showing that SCRs in the complex domain can approximate, to arbitrary precision, any unrestricted linear reservoir with a non-linear readout. We furthermore introduce a novel method to drastically reduce the number of SCR units, making such highly constrained architectures natural candidates for low-complexity hardware implementations. Our findings are supported by empirical studies on real-world time series datasets.
Original languageEnglish
Title of host publicationAAAI-25 Technical Tracks 16
Subtitle of host publicationThirty-Ninth AAAI Conference on Artificial Intelligence, Thirty-Seventh Conference on Innovative Applications of Artificial Intelligence, Fifteenth Symposium on Educational Advances in Artificial Intelligence
EditorsToby Walsh, Julie Shah, Zico Kolter
Place of PublicationWashington, DC
PublisherAAAI Press
Pages16622-16629
Number of pages8
Edition1
ISBN (Print)9781577358978
DOIs
Publication statusPublished - 11 Apr 2025
Event39th AAAI Conference on Artificial Intelligence - Philadelphia Marriott Downtown, Philadelphia, United States
Duration: 25 Feb 2025 → 4 Mar 2025
https://aaai.org/conference/aaai/aaai-25/

Publication series

NameProceedings of the AAAI Conference on Artificial Intelligence
PublisherAAAI
Number16
Volume39
ISSN (Print)2374-3468
ISSN (Electronic)2374-3468

Conference

Conference39th AAAI Conference on Artificial Intelligence
Abbreviated titleAAAI-2025
Country/TerritoryUnited States
CityPhiladelphia
Period25/02/25 → 4/03/25
Internet address

Fingerprint

Dive into the research topics of 'Universality of Real Minimal Complexity Reservoir'. Together they form a unique fingerprint.

Cite this