ML Anthology

Effect of Spatial Pooler Initialization on Column Activity in Hierarchical Temporal Memory

Mackenzie Leake, Liyu Xia, Kamil Rocki, Wayne Imaino
AAAI 2015 pp. 4176-4177
doi:10.1609/AAAI.V29I1.9734 /aaai/2015/leake2015aaai-effect/

Abstract

In the Hierarchical Temporal Memory (HTM) paradigm the effect of overlap between inputs on the activation of columns in the spatial pooler is studied. Numerical results suggest that similar inputs are represented by similar sets of columns and dissimilar inputs are represented by dissimilar sets of columns. It is shown that the spatial pooler produces these results under certain conditions for the connectivity and proximal thresholds at initialization. Qualitative arguments about the learning dynamics of the spatial pooler are then discussed.

PDF AAAI Semantic Scholar

Cite

Text

Leake et al. "Effect of Spatial Pooler Initialization on Column Activity in Hierarchical Temporal Memory." AAAI Conference on Artificial Intelligence, 2015. doi:10.1609/AAAI.V29I1.9734

Markdown

[Leake et al. "Effect of Spatial Pooler Initialization on Column Activity in Hierarchical Temporal Memory." AAAI Conference on Artificial Intelligence, 2015.](https://mlanthology.org/aaai/2015/leake2015aaai-effect/) doi:10.1609/AAAI.V29I1.9734

BibTeX

@inproceedings{leake2015aaai-effect,
  title     = {{Effect of Spatial Pooler Initialization on Column Activity in Hierarchical Temporal Memory}},
  author    = {Leake, Mackenzie and Xia, Liyu and Rocki, Kamil and Imaino, Wayne},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2015},
  pages     = {4176-4177},
  doi       = {10.1609/AAAI.V29I1.9734},
  url       = {https://mlanthology.org/aaai/2015/leake2015aaai-effect/}
}