1/1
Supplementary Material for "Improving Robot Motor Learning with Negatively Valenced Reinforcement Signals"
Download all (3.79 MB) This item is shared privately
dataset
modified on 2017-03-29, 09:19 The Training and Testing sets were used to train, test the solutions of the genetic algorithm used for hyperparameter optimisation. The Validation set was used to run 10 random initializations for the best 4 hyperparameter sets.
Results for the 10 random initializations of the 4 best hyperparameter sets, these files were used to produce the results of the tables 3, 4 and 5.
Average result per metric for the 4 best hyperparameters sets per condition. These results were used to produce figure 4, 5 and 6.
Commands used to compute two-ways ANOVA and multiple comparisons in Matlab 2015a.
All figures included in the paper and supplementary material document as zip files.
Reference
Navarro-Guerrero, N., Lowe, R., & Wermter, S. (2017). Improving Robot Motor Learning with Negatively Valenced Reinforcement Signals. Frontiers in Neurorobotics, 11(10). https://doi.org/10.3389/fnbot.2017.00010
Contact:
For information please see the paper. For specific questions regarding the paper please contact Nicolás Navarro-Guerrero .
Results for the 10 random initializations of the 4 best hyperparameter sets, these files were used to produce the results of the tables 3, 4 and 5.
Average result per metric for the 4 best hyperparameters sets per condition. These results were used to produce figure 4, 5 and 6.
Commands used to compute two-ways ANOVA and multiple comparisons in Matlab 2015a.
All figures included in the paper and supplementary material document as zip files.
Reference
Navarro-Guerrero, N., Lowe, R., & Wermter, S. (2017). Improving Robot Motor Learning with Negatively Valenced Reinforcement Signals. Frontiers in Neurorobotics, 11(10). https://doi.org/10.3389/fnbot.2017.00010
Contact:
For information please see the paper. For specific questions regarding the paper please contact Nicolás Navarro-Guerrero .
