Unsupervised Feature Learning Improves Prediction of Human Brain Activity in Response to Natural Images
The decoding performance was defined as the accuracy of identifying the 120 images in the validation set from a set of 9264 candidate images. The decoding performance of the SC2 model was significantly higher than that of the GWP2 model. Error bars show 1 SEM across the images in the validation set (bootstrapping method). A more detailed figure that shows the identified images is provided at http://www.ccnlab.net/research/.