Name
Abstract | ||
|---|---|---|
We present RNNbow, an interactive tool for visualizing the gradient flow during backpropagation in training of recurrent neural networks. By visualizing the gradient, as opposed to activations, RNNbow offers insight into how the network is learning. We show how it illustrates the vanishing gradient and the training process. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/MCG.2018.2878902 | IEEE Computer Graphics and Applications |
Keywords | Field | DocType |
Training,Recurrent neural networks,Visualization,Tools,Data visualization,Backpropagation,Bars | Computer vision,Data visualization,Computer science,Visualization,Recurrent neural network,Artificial intelligence,Backpropagation | Journal |
Volume | Issue | ISSN |
38 | 6 | 0272-1716 |
Citations | PageRank | References |
4 | 0.37 | 0 |
Authors | ||
6 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Dylan Cashman | 1 | 10 | 3.11 |
| g patterson | 2 | 318 | 11.83 |
| Abigail Mosca | 3 | 12 | 2.14 |
| Nathan Watts | 4 | 4 | 0.37 |
| Shannon Robinson | 5 | 4 | 0.37 |
| Remco Chang | 6 | 983 | 64.96 |