Unsupervised Learning in Space and Time by Marius Leordeanu

Author:Marius Leordeanu
Language: eng
Format: epub
ISBN: 9783030421281
Publisher: Springer International Publishing

4.5.2 Varying the Amount of Unsupervised Data

We also evaluated the influence on the performance of varying amounts of unsupervised (unlabeled) data used for unsupervised learning. We present the following experimental setup: first, we randomly split the unlabeled test data into two equal sets of frames, and . We have used one set of unlabeled frames for unsupervised learning and the other set for testing. While keeping the test frames set constant, and 8 labeled video shots per class for feature sign estimation from the training set (with 10 evenly spaced frames per shot), we varied the amount of unlabeled data used for unsupervised learning, by varying the percentage of unlabeled frames used from . We present results over 30 random runs in Fig. 4.14 and Tables 4.4 and 4.5.

Fig. 4.14Testing accuracy vs. amount of unlabeled data used for unsupervised learning. Note that the performance almost reaches a plateau after a relatively small fraction of unlabeled frames are added

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.