Interesting.
What I’m hearing is that while the original two methods likely achieve the
requisite attention spreading they don’t do this fast enough.
Rather than introduce a new third method, it seems to me that we could simply
replace the spreading of importance from everything in the AF
Ahh — I was thinking that the default setting for K would be one but that this
would change for specific contexts.
> On Sep 12, 2016, at 10:59 AM, Ben Goertzel wrote:
>
>> Rather than introduce a new third method, it seems to me that we could
>> simply replace the spreading
This does appear interesting.
> On Dec 11, 2016, at 9:09 AM, Ben Goertzel wrote:
>
> Wow, we really need to collaborate with these guys, this is important
> and promising stuff and OpenCog has a lot to contribute... well it
> does as soon as Nil and I solve scalable
Yes all of this makes a lot of sense and could be exciting, especially the
connection to Phi. Could be an extremely interesting paper — but a lot will
depend upon parameter tuning as well as creating an appropriate experiment.
Before we begin setting up the experiment to look for connections
ct attractor formations within ECAN dynamics prior
>> to looking at the larger ECAN dynamics using PLSI to reduce the space.
>>
>> Just trying to create a simple first set of experiments. Thoughts?
>>
>> --matt
>>
>> Sent from my iPhone
>>
>> On Ma
This is straightforward: Strength is a measure of likelihood — it can be
thought of as a probability, while confidence is a measure of how confident one
is in the strength value. Confidence is related to the value of count. The more
pieces of evidence upon which the strength is determined, the
Hi,
In helping out in modeling network growth of SIngularityNET, I have need for
any data (quantitative or qualitative) concerning the success of our own
algorithms compared to alternatives.
I know our algorithms are unique and more diverse. I also have some data
comparing PLN+ECAN to MLNs,
Hmmm. The code snippet I saw looks nothing like what the PLN deduction rule
should look like, but rather a heuristic that binarizes the result perhaps for
efficiency or for some specialized use case. As Linas says, Nil might know more
about why that deduction rule is coded the way it is. The
> On Jul 29, 2020, at 10:39 AM, Ben Goertzel wrote:
>
> On Wed, Jul 29, 2020 at 6:35 AM Abdulrahman Semrie wrote:
>>
>>> I think it's a mistake to try to think of a distributed atomspace as one
>>> super-giant, universe-filling uniform, undifferentiated blob of storage.
>>
>> It is not