Mercurial > ift6266

comparison writeup/aistats_review_response.txt @ 623:d44c78c90669

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
entered revisions for AMT and SVMs
author Yoshua Bengio <bengioy@iro.umontreal.ca>
date Sun, 09 Jan 2011 22:00:39 -0500
parents 5c67f674d724
children 49933073590c
comparison
equal deleted inserted replaced
622:09b7dee216f4 623:d44c78c90669
26 Linear SVM, NISTP, 800k, original, 88.50%, 85.24%, 87.36% 26 Linear SVM, NISTP, 800k, original, 88.50%, 85.24%, 87.36%
27 Linear SVM, NISTP, 800k, sparse quadratic, 81.76%, 83.69%, 85.56% 27 Linear SVM, NISTP, 800k, sparse quadratic, 81.76%, 83.69%, 85.56%
28 RBF SVM, NISTP, 100k, original, 74.73%, 56.57%, 64.22% 28 RBF SVM, NISTP, 100k, original, 74.73%, 56.57%, 64.22%
29 29
30 The best results were obtained with the sparse quadratic input features, and 30 The best results were obtained with the sparse quadratic input features, and
31 training on the CLEAN data (NIST) rather than the perturbed data (NISTP). 31 training on the CLEAN data (NIST) rather than the perturbed data (NISTP).
32 A summary of the above results was added to the revised paper.
32 33
33 34
34 * Using distorted characters as the corruption process of the Denoising 35 * Using distorted characters as the corruption process of the Denoising
35 Auto-Encoder (DAE). We had already performed preliminary experiments with this idea 36 Auto-Encoder (DAE). We had already performed preliminary experiments with this idea
36 and it did not work very well (in fact it depends on the kind of distortion 37 and it did not work very well (in fact it depends on the kind of distortion
57 complete (10 predictions) (3) discarding responses for which for which the 58 complete (10 predictions) (3) discarding responses for which for which the
58 time to predict was smaller than 3 seconds for NIST (the mean response time 59 time to predict was smaller than 3 seconds for NIST (the mean response time
59 was 20 seconds) and 6 seconds seconds for NISTP (average response time of 60 was 20 seconds) and 6 seconds seconds for NISTP (average response time of
60 45 seconds) (4) discarding responses which were obviously wrong (10 61 45 seconds) (4) discarding responses which were obviously wrong (10
61 identical ones, or "12345..."). Overall, after such filtering, we kept 62 identical ones, or "12345..."). Overall, after such filtering, we kept
62 approximately 95% of the AMT workers' responses. We thank the reviewer for 63 approximately 95% of the AMT workers' responses. The above paragraph
64 was added to the revision. We thank the reviewer for
63 the suggestion about multi-stage questionnaires, we will definitely 65 the suggestion about multi-stage questionnaires, we will definitely
64 consider this as an option next time we perform this experiment. However, 66 consider this as an option next time we perform this experiment. However,
65 to be fair, if we were to do so, we should also consider the same 67 to be fair, if we were to do so, we should also consider the same
66 multi-stage decision process for the machine learning algorithms as well. 68 multi-stage decision process for the machine learning algorithms as well.
67 69