It has been a while since my last post and in the intervening time I have been busy working on the code of my previous few posts.
During the course of this I have noticed that there are some further improvements to be made in terms of robustness etc. inspired by this Master's thesis, Improved Learning Algorithms for Restricted Boltzmann Machines, by KyungHyun Cho. Using the Deepmat Toolbox code available here as a guide, I now intend to further improve my code by incorporating the concepts of Parallel Tempering and adaptive learning rates for both the RBM and CRBM training.
More in due course.
