RC-SFA

Benchmark: 
Place cell learning
Result: 
17.2

 Using the Place cell reconstruction method mentioned in the Benchmark description and a hierarchical RC-SFA architecture similar to [1], the optimal result was an test error of 17.2 cm between true robot position and predicted robot position. 

Results are to be submitted for a journal.

References

  1. Antonelo, E A., B. Schrauwen, and D. Stroobandt,
    "Unsupervised Learning in Reservoir Computing: Modeling Hippocampal Place Cells for Small Mobile Robots",
    International Conference on Artificial Neural Networks (ICANN), 2009.
Related Publications: 
Unsupervised Learning in Reservoir Computing: Modeling Hippocampal Place Cells for Small Mobile Robots
Posted on Wed, 31/03/2010 - 14:35 by admin