Evolving the Hierarchical Temporal Memory using Genetic Algorithm



Hi All,

I have worked on the idea of evolving HTM using genetic algorithm on the problem of anomaly detection and evaluate the HTM performance using NAB by its score. I have also evaluate different parameters of HTM and observed critical parameters by feature selection (Pearson’s Correlation). I have completed and presented this work successfully in masters thesis. Now I want to publish this work. I am planning to submit my article in neurocomputing journal.

Can anyone in the community guide me or if anyone have relevant experience on publication and interest in NAB and HTM (java) then I can work on article with collaboration. I want to submit my work as quickly as possible.

I have seen an article of @scott. Can you please guide me or have a look on my work ?



I will help review your work if you share it. My only advice about publishing is to choose an open access journal so more people can see it without having to pay for it.


What do you mean by “genetic algorithm”? Which part of HTM are you evolving?

I’ve done some research using TPE to optimise model parameters here: https://github.com/JonnoFTW/htm-models-adelaide/blob/master/engine/sm_model/optimize_htm.py