Deep Learning and System Identification⁎This research was financially supported by the Swedish Foundation for Strategic Research (SSF) via the project ASSEMBLE (contract number: RIT15-0012) and by the Swedish Research Council via the projects Learning flexible models for nonlinear dynamics (contract number: 2017-03807) and NewLEADS - New Directions in Learning Dynamical Systems (contract number: 621-2016-06079). Ljung's work was supported by Vinnova's Competence Center LinkSic. Issue 2 (2020)

• Record Type: Journal Article
• Authors: Ljung, Lennart; Andersson, Carl; Tiels, Koen; Schön, Thomas B.
• Is Part Of: IFAC-PapersOnLine. Volume 53:Issue 2(2020)
• Page Start: 1175
• Page End: 1181
• Publication Date: 2020
• Subjects: Model structure -- Bias/Variance Trade-off -- Model Validation -- LSTM -- Cascadeforwardnet -- Deep nets; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2020.12.1329
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: Deep learning is a topic of considerable interest today. Since it deals with estimating - or learning - models, there are connections to the area of System Identification developed in the Automatic Control community. Such connections are explored and exploited in this contribution. It is stressed that common deep nets such as feedforward and cascadeforward nets are nonlinear ARX (NARX) models, and can thus be easily incorporated in System Identification code and practice. The case of LSTM nets is an example of NonLinear State-Space (NLSS) models.