Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Nov 1, 2010
With each method we learn a 4 linear classfiers (SVM) that predicts whether we should press each button (left,right,speed,jump) given the current image features (each image is discretized in a 22x22 grid centered around mario, 14 binary features describe the content of each cell, and we predict given a history of these features over the last 4 images)