Re: [Idnet] IETF99 for applying AI/ML into network management: Follow-up

[Pedro Martinez-Julia <[e-mail]>]

Dear Yanshen,

Please find my reply below.

On Fri, Jul 21, 2017 at 08:52:33AM +0000, yanshen wrote:
> Dear Pedro,
> 
> Thanks for your hints. Please find some hints from my side.
> 
> I am not sure whether you means that there should be an "interface" or
> an "entity" between the other source data and control loop?

I'm sorry, I mean that, by definition, a "closed control loop" intends
to maintain a particular variable within some defined values by reading
it and enforcing the necessary actions to affect it. Any external data
cannot be considered as such variable so they cannot be part of the
closed control loop. However, it can be used to help controlling other
variables, but it must be clearly defined how the external data is
integrated and derived into the changes to the other variable.

> BTW, have you ever obtain some achievement about such point that can
> be shared?

My results are trivial and do not shed any light to this issue and I do
not try to fit external data into the closed control loop, I just use it
to complement the control of the "load" variable.

> I think that since NMRG may be a probably choice, it is better to
> explore via this thread. 

So let's go that way.

> Thanks for reading. 
> 
> Regards,
> 
> Yansen

Regards,
Pedro

-- 
Pedro Martinez-Julia
Network Science and Convergence Device Technology Laboratory
Network System Research Institute
National Institute of Information and Communications Technology (NICT)
4-2-1, Nukui-Kitamachi, Koganei, Tokyo 184-8795, Japan
Email: [e-mail]
---------------------------------------------------------
*** Entia non sunt multiplicanda praeter necessitatem ***