Real-Time Recurrent Learning
GPTKB entity
Statements (17)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:algorithm
|
| gptkbp:abbreviation |
RTRL
|
| gptkbp:complexity |
O(n^4) per time step
|
| gptkbp:computes |
exact gradients for RNNs
|
| gptkbp:contrastsWith |
Backpropagation Through Time
|
| gptkbp:field |
gptkb:artificial_intelligence
gptkb:machine_learning neural networks |
| https://www.w3.org/2000/01/rdf-schema#label |
Real-Time Recurrent Learning
|
| gptkbp:introducedIn |
1989
|
| gptkbp:limitation |
computationally expensive
|
| gptkbp:proposedBy |
gptkb:Ronald_J._Williams
David Zipser |
| gptkbp:updated |
weights online
|
| gptkbp:usedFor |
training recurrent neural networks
|
| gptkbp:bfsParent |
gptkb:BPTT
|
| gptkbp:bfsLayer |
8
|