Research Article
Skill Learning for Intelligent Robot by Perception-Action Integration: A View from Hierarchical Temporal Memory
Table 2
Main optimal parameters for SP and TM algorithms.
| | Parameters | Description | Value |
| | Threshold for the number of active synapses on a segment | 15 | | Threshold for the permanence of potential synapse | 0.2 | | Initial value of the boost factor | 1.0 | | Maximal boost factor | 2.0 | | Initial value of the inhibition radius | 0 | | Minimal number of winning columns | 1 | | Incremented permanence value in spatial pooling | 0.05 | | Decremented permanence value in spatial pooling | 0.05 | | Any synapse whose permanence value is above this threshold will become an active synapse | 0.1 | | Minimum active duty cycle | 0.001 | | Threshold used to determine whether a distal segment is activated | 14 | | Incremented permanence value in temporal pooling | 0.1 | | Decremented permanence value in temporal pooling | 0.1 |
|
|
