Factored Conditional Restricted Boltzmann Machines for Modeling ...

Version 0.100 (Unsupported, unreleased) Code provided by Graham Taylor and Geoff Hinton For more information, see: http://www.cs.toronto.edu/~gwtaylor/publications/icml2009 Permission is granted for anyone to copy, use, modify, or distribute this program and accompanying programs and documents for any purpose, provided this copyright notice is retained and prominently displayed, along with a note saying that the original programs are available from our web page. The programs and documents are distributed without any warranty, expressed or implied. As the programs were written for research purposes only, they have not been tested to the degree that would be advisable in any important application. All use of these programs is entirely at the user's own risk. This subdirectory contains files related to learning and generation: demo_train.m Main file for training a model demo_generate.m Main file for generating from a trained model Note that, by default, we train a factored conditional RBM with three-way interactions whose feature-factor weights are shared between the three "sub-models" (see the paper for the definition of sub-models). The following learning and generation scrips are called by the demos: gaussianfbm_sharefeatfac.m Trains the model (shared feature-factor weights) gen_sharefeatfac.m Generates from a trained model (shared feature-factor weights) However, there are other weight-sharing options. Some of these are also implemented. They have different learning and generation scripts: No sharing: gaussianfbm.m, gen.m Sharing feature-factor, past-factor weights: gaussianfbm_sharefpfac.m, gen_sharefpfac.m Sharing feature-factor, past-factor, visible-factor weights: gaussianfbm_sharefpvfac.m, gen_sharefpvfac.m Sharing feature-factor, past-factor, visible-factor, hidden-factor weights (full-sharing): gaussianfbm_shared.m, gen_shared.m Sharing feature-factor, past-factor, visible-factor weights: gaussian Also included are learning and generation scripts for a factored version of the CRBM that does not have any three-way interactions: gaussianfaccrbm.m, genfaccrbm.m The Motion subdirectory contains files related to motion capture data: preprocessing/postprocessing, playback, etc ... Acknowledgments The sample data we have included has been provided by the CMU Graphics Lab Motion Capture Database: http://mocap.cs.cmu.edu Several subroutines related to motion playback are adapted from Neil Lawrence's Motion Capture Toolbox: http://www.cs.man.ac.uk/~neill/mocap/ Several subroutines related to conversion to/from exponential map representation are provided by Hao Zhang: http://www.cs.berkeley.edu/~nhz/software/rotations/