Development and Analysis of non-standard Echo State Networks
Seiten
2024
TUDpress (Verlag)
978-3-95908-648-6 (ISBN)
TUDpress (Verlag)
978-3-95908-648-6 (ISBN)
In an era of complex deep learning architectures like transformers, CNNs, and LSTM cells, the challenge persists: the hunger for labeled data and high energy. This dissertation explores Echo State Network (ESN), an RNN variant. ESN‘s efficiency in linear regression training and simplicity suggest pathways to resource-efficient, adaptable deep learning. Systematically deconstructing ESN architecture into flexible modules, it introduces basic ESN models with random weights and efficient deterministic ESN models as baselines. Diverse unsupervised pre-training methods for ESN components are evaluated against these baselines. Rigorous benchmarking across datasets — time-series classification, audio recognition — shows competitive performance of ESN models with state-of-the-art approaches. Identified nuanced use cases guiding model preferences and limitations in training methods highlight the importance of proposed ESN models in bridging reservoir computing and deep learning.
Erscheinungsdatum | 16.02.2024 |
---|---|
Sprache | englisch |
Maße | 175 x 245 mm |
Themenwelt | Technik ► Elektrotechnik / Energietechnik |
Schlagworte | Analysis • Echo • Networks |
ISBN-10 | 3-95908-648-2 / 3959086482 |
ISBN-13 | 978-3-95908-648-6 / 9783959086486 |
Zustand | Neuware |
Haben Sie eine Frage zum Produkt? |
Mehr entdecken
aus dem Bereich
aus dem Bereich
DIN-Normen und Technische Regeln für die Elektroinstallation
Buch | Softcover (2023)
Beuth (Verlag)
CHF 119,95
Kolbenmaschinen - Strömungsmaschinen - Kraftwerke
Buch | Hardcover (2023)
Hanser (Verlag)
CHF 69,95