Can Neural Networks Dream?
Abstract
According to Crick and Mitchison’s theory of reorganization, the reverse learning mechanism is responsible for preventing the brain from getting overloaded. This mechanism occurs in (rapid eye movement) REM sleep and is closely associated with dreams. In this study, the reverse learning mechanism is applied to artificial neural networks (ANNs), and its effects on catastrophic investigated. The network’s performance with respect to catastrophic forgetting is compared before and after applying the reverse learning mechanism.
Fields of Study & Subjects
Domain Adaptation and Few-Shot Learning
Neural Networks and Applications
Sleep and Wakefulness Research
Publication details
- Journal / source
- CRC Press eBooks
- Pages
- 63–88
- Publication date
- 27 February 2026
- Language
- EN
- Document type
- Book chapter