Definitely a valid question.

I've had success <http://mrcslws.com/gorilla/?path=hotgym.clj>
hand-crafting model parameters. I wrote that whole post without running a
swarm, except to compare final results.

Speaking from my experience, I think swarming could be useful to use
side-by-side with "intelligent design". As I wrote that post, I wanted a
way to randomly explore the parameter-space around my current solution.
Then, ideally, I'd investigate why that parameter-change causes an
improvement, and then make a decision about what to do next. So I'd be
using reasoning, but part of my reasoning is knowing that my current mental
model is probably wrong, and random exploration is a good heuristic for
breaking through that.

There are other answers to your question, corresponding to other NuPIC
audiences. Not everyone wants to dive right into Hard Mode. :)

On Fri, Dec 4, 2015 at 12:28 PM, Ryan Singer <[email protected]> wrote:

> Hello NuPIC community,
>
> I'm just getting started. I'm excited about Nupic because the HTM model
> and SDR data structure make so much sense intuitively. Finally, an approach
> that doesn't feel arbitrary and over-engineered.
>
> However I'm confused about why swarming is necessary when configuring
> models. I've been reading all the docs and I haven't yet found an
> explanation of why a human can't arrive at the right model parameters
> through reasoning.
>
> Am I missing some documentation or video that explains this? Any help is
> appreciated.
>
> Ryan
>

Reply via email to