Mercurial > cortex

comparison thesis/cortex.org @ 549:c14545acdfba

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
minor edit.
author Robert McIntyre <rlm@mit.edu>
date Fri, 02 May 2014 03:34:13 -0400
parents 0b891e0dd809
children b1d8d9b4b569
comparison
equal deleted inserted replaced
548:0b891e0dd809 549:c14545acdfba
3207 #+end_src 3207 #+end_src
3208 #+end_listing 3208 #+end_listing
3209 3209
3210 After about 1 minute of manual training, I was able to achieve 95% 3210 After about 1 minute of manual training, I was able to achieve 95%
3211 accuracy on manual testing of the worm using =init-interactive= and 3211 accuracy on manual testing of the worm using =init-interactive= and
3212 =test-empathy-accuracy=. The majority of errors are near the 3212 =test-empathy-accuracy=. The majority of disagreements are near the
3213 boundaries of transitioning from one type of action to another. 3213 transition boundaries from one type of action to another. During
3214 During these transitions the exact label for the action is more open 3214 these transitions the exact label for the action is often unclear,
3215 to interpretation, and disagreement between empathy and experience 3215 and disagreement between empathy and experience is practically
3216 is essentially irrelevant at this point, giving a practical 3216 irrelevant. Thus, the system's effective identification accuracy is
3217 identification accuracy of even higher than 95%. When I watch this 3217 even higher than 95%. When I watch this system myself, I generally
3218 system myself, I generally see no errors in action identification. 3218 see no errors in action identification compared to my own judgment
3219 of what the worm is doing.
3219 3220
3220 ** Digression: Learning touch sensor layout through free play 3221 ** Digression: Learning touch sensor layout through free play
3221 3222
3222 In the previous section I showed how to compute actions in terms of 3223 In the previous section I showed how to compute actions in terms of
3223 body-centered predicates, but some of those predicates relied on 3224 body-centered predicates, but some of those predicates relied on