Name
Title | ||
|---|---|---|
Discretization of ISO-Learning and ICO-Learning to Be Included into Reactive Neural Networks for a Robotics Simulator |
Abstract | ||
|---|---|---|
Isotropic Sequence Order learning (ISO-learning) and Input Correlation Only learning (ICO-learning) are unsupervised neural algorithms to learn temporal differences. The use of devices implementing this algorithms by simulation in reactive neural networks is proposed. We have applied several modifications to original rules: weights sign restriction, to adequate ISO-learning and ICO-learning devices outputs to the usually predefined kinds of connections (excitatory/inhibitory) used in neural networks, and decay term inclusion for weights stabilization. Original experiments with these algorithms are replicated as accurate as possible with a simulated robot and a discretization of the algorithms. Results are similar to those obtained in original experiments with analogue devices. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1007/978-3-540-73055-2_39 | IWINAC (2) |
Keywords | Field | DocType |
reactive neural networks,neural algorithm,weights sign restriction,neural network,ico-learning devices output,robotics simulator,reactive neural network,weights stabilization,original rule,isotropic sequence order learning,adequate iso-learning,original experiment,temporal difference | Discretization,Computer science,Robotics simulator,Hebbian theory,Artificial intelligence,Behavior-based robotics,Artificial neural network,Robot,Neural algorithms,Machine learning,ICO | Conference |
Volume | ISSN | Citations |
4528 | 0302-9743 | 2 |
PageRank | References | Authors |
0.39 | 7 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| José Manuel Cuadra Troncoso | 1 | 53 | 10.58 |
| José R. Álvarez Sánchez | 2 | 3 | 0.76 |
| Félix Paz López | 3 | 2 | 0.39 |