Open Access
Table 2.
Hyperparameter setup for the mixed-input multilayer perceptron θNN ≡ θMLP used as our deep survival network.
| Hyperparameter | Value |
|---|---|
| (Hidden) layers & nodes | 64, 32, 16, 8, 4, 2 |
| Learning rate | 0.0001 |
| Dropout | 0.1 |
| Batch size | 150 |
| Decoupled weight decay | 0.01 |
| Cycle-η-multiplier | 0.8 |
| Cycle-multiplier | 2 |
Notes. The second cell provides the parameter setup for the AdamWR optimizer (Loshchilov & Hutter 2017a) used inside θMLP.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.