Are there any plans for providing free Neuroscience courses by Numenta something like HTM school, something with a little bit of structure? I would imagine this as tailored for Numenta’s research interests for example cracking cortical algorithms. As an admirer of Numenta’s work, having a computing background without confidence in Neuroscience is very limiting. And most of the time when people talk about the cortex in relation to some algorithm, there is no common/default reference, which may cause a convoluted discussion.
I believe having these courses if there are none yet, would help a lot in educating htm/cortical-based ML practitioners whatever you may wanna call us. It may also help/remind everyone about Numenta’s strategy to attaining MI. also Just like HTM school, we can refer ro these courses confidently and have a common baseline understanding.
Just my thoughts. A lot of NS info is out there in the web I know but scraping and reorganiZing them to learn fundamentals of NS that will be useful for Numenta-based ML or research is a lot of work and can be easily become a time-waster.
Thanks, do you think a more tailored course for Numenta would be much more useful? Like HTM school there are references to the actual application in HTM during discussion which I found very useful.
I’m happy to see that there is a demand for this. You fit my target audience which is the AI community interested to get insights from brain mechanisms.
I intend to have two main parts:
The first part is about some neurosciences theories & mechanisms that may be useful from an AI researcher’s point of view.
The second part looks at some today’s neuroscience-grounded AI approaches (main ideas from Numenta included of course!)
I’ll do my best to finish a complete draft in January
Great to hear this.
Once the draft is complete, I would suggest that we publish it as a Wiki, but maybe people’s edits might dilute it.
I think that BAMI is along similar lines. https://numenta.com/assets/pdf/biological-and-machine-intelligence/BAMI-Complete.pdf
Along these lines, I tried to create my own Mendeley site, but see that others are better organized.
That is the difference in research, it is not “canonized knowledge”, so subject to updates and interpretation.
Thanks for the idea, Jose! We don’t have any plans on doing this. It sounds like a pretty big job, and not one that I’m necessarily qualified for. I’m looking forward to anything @mthiboust creates! I am still going to be working on BHTMS this coming year, however (hopefully it will eventually replace BAMI). I hope to add some more neuroscience to this resource, but it has not been my top priority lately.
My ebook is nearly finished, but the last mile is taking more time than expected
I am currently taking into account reviews from a few people (including @Bitking, @Casey and @Falco from this HTM community and paper/book authors I cited)
I hope to release it this month to give some reading during the social distancing measures in most countries!
