Hello everyone,
I’m having some trouble understanding Figure 5 from “The HTM Spatial Pooler” by Cui et. al (2017) and was hoping to get some clarifications
First, I wanted to check if the above image (Fig. 5A) is representing the input space or the minicolumns. I assume it is the minicolumns as it is 32 x 32, but the color bar says the “number of connected column”, which could be interpreted as the number of minicolumns each input is connected to
I was also confused by Fig. 5 C, as the axis are labeled by 32 x 32, but there more than five dots inbetween the ticks. Are these dots representing the receptive field in a large input space (such as a large MNIST image) squeezed into a 32 x 32 region? A similar question was asked in (What does "the receptive field center" mean in SP? - #2 by Bitking)
Thanks for your help and let me know if I can clarify my questions