Bidirectional Associative Memories: Unsupervised Hebbian Learning to Bidirectional Backpropagation
| Autor: | Bart Kosko |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0209 industrial biotechnology
Quantitative Biology::Neurons and Cognition business.industry Computer science Competitive learning Supervised learning 02 engineering and technology Content-addressable memory Backpropagation Computer Science Applications Human-Computer Interaction Synapse Matrix (mathematics) 020901 industrial engineering & automation Hebbian theory Control and Systems Engineering 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence Electrical and Electronic Engineering Layer (object-oriented design) business Software Associative property |
| Zdroj: | IEEE Transactions on Systems, Man, and Cybernetics: Systems. 51:103-115 |
| ISSN: | 2168-2232 2168-2216 |
| DOI: | 10.1109/tsmc.2020.3043249 |
| Popis: | Bidirectional associative memories (BAMs) pass neural signals forward and backward through the same web of synapses. Earlier BAMs had no hidden neurons and did not use supervised learning. They tuned their synaptic weights with unsupervised Hebbian or competitive learning. Two-layer feedback BAMs always converge to fixed-point equilibria for threshold or threshold-like neurons. Every rectangular connection matrix is bidirectionally stable. These simpler BAMs extend to arbitrary hidden layers with supervised learning if the resulting bidirectional backpropagation algorithm uses the proper layer likelihood in the forward and backward directions. Bidirectional backpropagation lets users run deep classifiers and regressors in reverse as well as forward. Bidirectional training exploits pattern and synaptic information that forward-only running ignores. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|