The don't do deep learning. Their papers are based on graphical models. Their RCN model can be said to be inspired by Numentas earlier HTM model that dileep developed there, but is very different in fundamental ways.
https://arxiv.org/abs/1611.02252
https://arxiv.org/abs/1611.02767
https://arxiv.org/abs/1706.04313
But it seems like lately they focus more on RL for robotics.