Table 2
Policy network training parameters and sampling configuration.
| Parameter | Value | Description | Sampling rate |
|---|---|---|---|
| State vector dimension | 12 | Combined geometric and topological indicators | Per step |
| Action frequency | 1 action/step | One action generated per step | Step-aligned |
| Reward weight α | 0.7 | Weight for geometric deviation | Fixed |
| Reward weight β | 0.3 | Weight for topological adjustment | Fixed |
| Batch size | 64 | Number of samples per training batch | Per iteration |
| Learning rate | 0.001 | Update rate for network parameters | Adaptive |
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.