Memory representation in moth olfactory network, role of neuromodulators


I have found two fascinating papers about modeling of moth olfactory network.
Model uses Hebbian plasticity and was calibrated on in vivo data.
Research includes model of mushroom body where sparsity and randomness play a critical role.
One of the interesting parts in model is the role of octopamine stimulation for learning.

What do you think about sparsity and memory representation in insects mushroom body?
Could you describe major differences in memory representation in neocortex and mushroom body?
Do you have experiments with neuromodulators in HTM model?


This paper expands on that and provides some more
computational/mathematical interpretation:

Sanjoy Dasgupta, Charles F. Stevens, Saket Navlakha – A neural
algorithm for a fundamental computing

key take aways

  • fly olfactory brain is an implementation of the HTM spatial pooler (my
  • mathematically it’s an implementation of dimensionality reduction by
    relying on locality guarantees provided by the Johnson-Lindenstrauss
    lemma that I mentioned before

This is the regular reminder that we still need a mathematical model of
the whole thing, or in other words pull a Kepler on the Copernicus.

– Rik