Relation between OpenCOG and "Expert System"

[Andrei Suiu]

What is the relation between OpenCOG and Expert System at this stage in 2017 ?

If OpenCOG is not basically an Expert System, does it exhibit all Expert System's properties?

I would be grateful if you Ben could provide an answer that elucidates this, as I've found some of your states comparing Watson and OpenCOG:

"Watson is basically a huge expert system with a knowledge base, fueled by information extraction."

"OpenCOG, once properly developed, will be a completely different sort of animal than Watson. Watson is ultimately a one-trick pony – and an architecture that can be used to create a series of other one-trick ponies, with a lot of human energy and innovation required for each. For instance, to make a Watson type system for the medical field, as IBM is now striving to do, is not mainly a matter of giving Watson medical data and having it learn. It involves a massive amount of domain-specific human effort. This is because Watson is not a general intelligence – it lacks the ability to generalize. But the whole point of an AGI architecture like OpenCog is that it will have the ability to achieve its goals in complex environments (like the real world) by GENERALIZING from its prior experience, rather than having to be specifically reprogrammed for each new kind of situation it has to deal with." (Source)

At the first view it looks like AtomSpace is basically a Knowledge Base, and the MindAgents are Inference Engines, that makes OpenCOG look like an Expert System according to this definition: "An expert system is divided into two subsystems: the inference engine and the knowledge base. The knowledge base represents facts and rules. The inference engine applies the rules to the known facts to deduce new facts" (Source)

Thanks,

   Andrei

[supahacka]

"For the AGI research community, a computer that could become a grandmaster at Go [...]" -- Ben

What another 4 years can do ...

[Ben Goertzel]

On Thu, Mar 9, 2017 at 2:03 AM, Andrei Suiu <andre...@gmail.com> wrote:
>
> At the first view it looks like AtomSpace is basically a Knowledge Base, and
> the MindAgents are Inference Engines, that makes OpenCOG look like an Expert
> System

No.

Most MindAgents are not inference engines -- e.g. calling clustering,
or ECAN (importance spreading), an "inference engine" is way too much
of a stretch

Further, "expert systems" typically have their knowledge bases filled
via curated or hand-coded knowledge, not via knowledge obtained by
learning

Of course, if you stretch the definition far enough you can convince
yourself a human baby is an "expert system" (it contains some expert
knowledge about drinking milk from its mommy, in its knowledge base),
but ... well ...



--
Ben Goertzel, PhD
http://goertzel.org

“Our first mothers and fathers … were endowed with intelligence; they
saw and instantly they could see far … they succeeded in knowing all
that there is in the world. When they looked, instantly they saw all
around them, and they contemplated in turn the arch of heaven and the
round face of the earth. … Great was their wisdom …. They were able to
know all....

But the Creator and the Maker did not hear this with pleasure. … ‘Are
they not by nature simple creatures of our making? Must they also be
gods? … What if they do not reproduce and multiply?’

Then the Heart of Heaven blew mist into their eyes, which clouded
their sight as when a mirror is breathed upon. Their eyes were covered
and they could see only what was close, only that was clear to them.”

— Popol Vuh (holy book of the ancient Mayas)

[Andrei Suiu]

So do I interpret the answer right that OpenCOG in fact is a non-typical Knowledge Base that in addition to the inference engines has some MindAgents that are able to evolve/extend knowledge using more than just inference(like clustering or importance spreading) ?

Of course, if you stretch the definition far enough you can convince
yourself a human baby is an "expert system"

--

Does this mean that definition of "Expert system" is too broad to be useful in differentiation between AI systems?

Thanks,

     Andrei

[Ben Goertzel]

Probably the concept of "expert system" is too old-school to be really useful

E.g. is Watson an expert system? Sort of. More so than OpenCog is,
for sure. OTOH Watson relies on all sorts of deep learning behind the
scenes, on natural language information extraction etc. etc. --
alongside some hard-coded rules. And Grady Booch has an AGI-oriented
cognitive architecture in mind, which is lurking somewhere inside the
Watson codebase...

[Michael S]

Hi sorry to reply to older thread, but could one make an expert system whose domain is learning itself (supervised or unsupervised)? In theory, I mean.

Like, such an expert system could be filled with knowledge about learning and epistemology, and its inference engine would be taught rules about how to learn that are very domain-general. Then perhaps, once you've laid the framework by teaching the system how to learn in a formal logical way as an expert system (as opposed to more "hidden" and not readily editable machine learning algorithms), I wonder if you could use that "learning expert" to learn other things better. I wonder if an expert system like this (or any other) could be used by a machine learning supersystem, or another expert supersystem--these would be higher-level intelligences then maybe.

I've personally been interested in approaching AGI from the formal logical perspective instead of the machine learning probabilistic perspective, so I thought this was interesting post.

[Linas Vepstas]

expert systems are unable to apply the knowledge they contain.  So for example, an expert system in geological formations is not actually able to drill oil wells ... it simply can provide yes/no answers to questions, or provide advice (e.g. "oil wells near the edges of granite domes are often productive")

Thus, an expert system in learning could maybe give advice to teachers about teaching, but that would not make it itself capable of learning ... the encoded knowledge is too abstract.  ("writing an essay after reading a book often leads to learning")

--linas

--
You received this message because you are subscribed to the Google Groups "opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencog+unsubscribe@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.

To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/53a4f7df-df60-4b70-9895-70f6ae72b9ca%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Michael S]

ah I see, thank you for explaining :)

To unsubscribe from this group and stop receiving emails from it, send an email to opencog+u...@googlegroups.com.

To post to this group, send email to ope...@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.