Skip to main content
. 2017 Dec 19;11:66. doi: 10.3389/fnbot.2017.00066

Table 4.

Experimental results of the CSL task for 25 and 40 trials.

Method ROW ARI_a ARI_p ARI_o ARI_c EAR_Fd
(A) 25 trials
Proposed 0 0.239 0.932 0.201 0.720 0.866
w/o MEC-II 0 0.299 0.971 0.207 0.717 0.723
mMLDA 0 0.255 0.959 0.226 0.703
Proposed 30 0.297 0.879 0.227 0.702 0.751
w/o MEC-II 30 0.242 0.980 0.218 0.683 0.601
mMLDA 30 0.296 0.893 0.256 0.730
Proposed 50 0.240 0.905 0.257 0.681 0.604
w/o MEC-II 50 0.224 0.895 0.211 0.694 0.482
mMLDA 50 0.221 0.981 0.303 0.688
(B) 40 trials
Proposed 0 0.304 0.960 0.240 0.691 0.988
w/o MEC-II 0 0.282 0.986 0.190 0.729 0.763
mMLDA 0 0.290 0.959 0.224 0.736
Proposed 30 0.303 0.978 0.193 0.698 0.829
w/o MEC-II 30 0.349 0.917 0.219 0.726 0.718
mMLDA 30 0.307 0.956 0.159 0.717
Proposed 50 0.316 0.922 0.199 0.718 0.668
w/o MEC-II 50 0.258 0.937 0.210 0.726 0.639
mMLDA 50 0.297 0.989 0.123 0.687

Bold and underscore indicate the highest evaluation values, and bold indicates the second highest evaluation values.