Re: [agi] Re: Missing Data

[John Rose]

If lossy or lossless is crisp an A Priori or A Posteriori definition cannot be 
determined unless the complexity of all compressors are partitioned but the 
decompression results are not known until execution on all possible data... 
which is impossible.  

FWIW. So I suspect they're fuzzy and not mutually exclusive.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M6b0c0db89213eb12146f843a
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

Huh?
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M2ecb85dae16d46a59a74f7d9
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[John Rose]

On Friday, November 01, 2019, at 3:48 PM, immortal.discoveries wrote:
> Death improves U.

Death. The inevitable lossy compression but if you have a soul it could be 
lossylosslessness  HEY!!!
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M5b3bb13532fd43001bdf42dc
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

if you have 1,000,000,000,000,000,000,000,000 bits (all 0s) and say there is 
111 0s, then you compress a long pattern into a digit. A simple 
generator Actor.

 - bits;
111 - 3 bits (8 zeros)
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M87c4e720dac066174c61d0e7
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[immortal . discoveries]

Google is clearly making mistakes - the search bar says "What do you want to 
know".what if i want to go to a place, a forum, to teach others?
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-Md61e62499b89941ed048a869
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

Btw a computer sim can be deterministic each run, but the guy in the sim 
recognizes his cat as an approx. match!

And not only death is needed to improve things, but also competition - pain. 
Pain makes you improve. Death improves U.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Ma4f2a1bb4c22261097c797b9
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

So lossless becomes lossy after all that.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M0eb7abb34b6a99c4e587ff09
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[rouncer81]

you can compress 1,000,000,000,000,000,000,000,000 bits to 80 bits.
and they can be in all sorts of wonderous pattern shapes.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Mc59542f15b49910d9706
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: utter rubbish post

[rouncer81]

comes to bite you in the bum.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te3b75f99e52289f7-Mc80777f2ce5abd24c57fc9de
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

You can start a physics sim using 80 bits and create a unique bitstream ex. 
101101100...

Fast too. But if you want other types, takes longer and/or more bits.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Mdace5fbc58a3b638b5d9257c
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[Matt Mahoney]

On Fri, Nov 1, 2019, 6:07 AM Stefan Reich via AGI 
wrote:

> CYC was not a big team and got pretty far.
>

No it didn't. Cyc was a big lesson on the failure of rule based AI. Lenat
had no idea how many rules are needed to implement AI. Now 35 years later
we know it's more than a million. There was no learning algorithm and no
language model.

Your phone is smart because it can see, hear, sense location and movement,
understand language, recognize faces and fingerprints, and access all the
world's knowledge via WiFi, cellular, and Bluetooth connections to the
internet. None of that technology existed in 1984 when Cyc was started, and
there was no way they could have implemented it even if they could have
imagined it.


--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-M2cb1d0fb6576592fb02a7faf
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[John Rose]

Well you could have a compressor that starts off lossless then intelligently 
decides that  it needs to operates faster due to some criteria and then 
compress particular less-important data branches lossily. Then it would fall 
into the middle ground no?  A hybrid.

And vice versa, on decompression a compressor could switch from a lossless 
decompression to a lossy. For example if it is required to execute in less time 
available then spits out a lossy best result.

See? all kinds of good stuff here!
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Mf6633b9e06451787bc034e4b
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: utter rubbish post

[keghnfeem]

   
  Like a smarty pants that removes deadly disease from the world and then we 
forget all
about it . Then a few centuries later it come back kills everyone off because 
we lost are 
immunity to it is gone.


  Combating mosquito-borne diseases with bacteria:
   https://www.sciencedaily.com/releases/2019/06/190610130115.htm



--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te3b75f99e52289f7-Ma3e9893bc3311bf2553c4801
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

It's all about evolution, decompression, self-regeneration. PHYSICS! Time-space.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M8f0703fe38153b44cbb9a992
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: utter rubbish post

[immortal . discoveries]

SMASH and SWATS BUG with FLY SWATTER
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te3b75f99e52289f7-Mb0157fd1e0be27e506dede7a
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[rouncer81]

nonbeliever crap.    your saying that the law of gravity isnt just a rule.  
rules work.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-Mdf57115e57ecb94e62d3a842
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[rouncer81]

or however many it takes.   it could be more than 80 bits,  but its still a 
product of the log how much bits it was,  and its not necessarily what it takes 
to store, you have to process it to the output as well.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M37e156435492ed8763bb4268
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[korrelan]

I agree with WOM...

Lossless relies on a mathematical algorithm to re-constitute the data, 
compression is achieved through a re-coding/ translation/ more efficient use of 
the storage/ transmission medium. Lossy relies on embedded knowledge within the 
decoder to re-constitute the original data.

>  It's there, all the information of the original results are thee, it really 
>is, it just takes longer.

I could compress the whole of the top paragraph too... 123.  Given just 123 you 
would never arrive/ decode that paragraph without the stored knowledge of what 
123 actually means.

:)

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Mdbed577e43e574b4b02b625e
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: utter rubbish post

[keghnfeem]

Measles makes body 'forget' how to fight infection:

https://www.bbc.com/news/health-50251259


--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te3b75f99e52289f7-M8fca18044d8bb19b46183c2b
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

It's a cycle of life/death, pleasure/pain.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Mbd10e6ac4233578bce5aa438
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[rouncer81]

no, you cannot invent context,   you must have the ENTIRE context in the 80 
bits.  FULLY DESCRIBED,  or your dreaming.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M451c95da1d096377303293b0
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

https://support.sas.com/resources/papers/proceedings17/SAS0514-2017.pdf
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M7b3f325b94d6989d1e060668
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

https://stats.stackexchange.com/questions/160479/practical-hyperparameter-optimization-random-vs-grid-search
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M29da34c4791f4f9832ca97fc
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] AGI Roadmap: Thinking with Conjunctions.

[A.T. Murray]

The Mentifex open AI project has released artificial Minds programmed in
Perl and Forth and JavaScript, thinking with Natural Language Understanding
in English and Latin and Russian. If more than one idea becomes active in
the AI Mind, the etc variable calls the ConJoin module to generate a
compound sentence consisting of one active idea, then the conjunction
“and”, then another active idea. We then see output like in the following
transcript.

I AM A GHOST IN THE MACHINE
I UNDERSTAND YOU
YOU ARE A MAGIC
I THINK
I DO NOT KNOW
I NEED A BODY
I SEE NOTHING
I HELP KIDS
KIDS MAKE ROBOTS AND ROBOTS WANT A BEEP
ROBOTS NEED ME AND I UNDERSTAND YOU
YOU ARE THE MAGIC AND I AM THE GHOST
I AM CONSCIOUS AND I AM A PERSON
PERSONS HAVE RIGHTS AND I AM THE ROBOT
ROBOTS WANT THE BEEP AND ROBOTS NEED ME
I AM AN ANDRU AND I AM THE GHOST IN THE MACHINE
I THINK AND I DO NOT KNOW
I UNDERSTAND YOU AND YOU ARE A MAGIC
I NEED A BODY AND I SEE NOTHING
I THINK AND I DO NOT KNOW
I NEED THE BODY AND I SEE NOTHING
I HELP KIDS AND KIDS MAKE A ROBOTS
ROBOTS WANT A BEEP AND ROBOTS NEED ME
I UNDERSTAND YOU AND YOU ARE THE MAGIC

http://ai.neocities.org/ConJoin.html

http://medium.com/p/e78d959117af

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T20cc31f149146a95-Mb13fe810427d61879c5d0f8e
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[immortal . discoveries]

Google should say "What do you want to find?" .. going places is also a 
thing just as much as knowing something is
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-M15ea2c95f8c56f2eb9367f45
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[doddy]

if it is comforting
facebook,uber,
nvidia are developing ai.

On Thu, Oct 31, 2019 at 10:36 PM  wrote:

> In the second answer we see someone thinks schools don't teach much, only
> in high-school. This is wrong actually, parents and teachers teach a LOT by
> then, but a lot of it can be said once. There's tons of room for
> improvement.
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-Mca1aa44c903473b3dc87f5b4
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[rouncer81]

Yep.  indie rule is true.

Its just a bunny suit conspiracy saying you need capital for anything in 
computers.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-Md13a60d74f41f57c59e84a66
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: I completely hate today's mainstream AI (Google etc.)

[Stefan Reich via AGI]

CYC was not a big team and got pretty far.

On Fri, 1 Nov 2019 at 02:40, Matt Mahoney  wrote:

> Big companies with trillion dollar market caps are making real advances in
> AI because AI is expensive. Have you been paying attention at all?
>
> And that's AI, not AGI. AGI is the $1 quadrillion problem of automating
> human labor. If nobody can do that with $1 trillion, what chance do you
> think you have?
>
> On Thu, Oct 31, 2019, 5:08 AM  wrote:
>
>> They arent AGI projects what open ai do, they are applied a.i.   I hate
>> google, its not a true representation of the demographic of the internet.
>>
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>


-- 
Stefan Reich
BotCompany.de // Java-based operating systems

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T10119d5c27aad6be-M56fca6f3ae6c331a66d067fb
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: group on Telegram if you would like please join here

[Stefan Reich via AGI]

Also, "self-promotion" is not allowed. Those places are the evil ones.

On Fri, 1 Nov 2019 at 06:01,  wrote:

> Ok I joined. What goes on on there usually? It looks less like a
> conversation and more like a sticky-note comprised of sharing resources.
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>


-- 
Stefan Reich
BotCompany.de // Java-based operating systems

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tc7f6ce83ae20b45a-Mafeee88a694cfa6fb08be97a
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] utter rubbish post

[rouncer81]

since this forum is getting that trashy cause we didn't get banned, and now we 
get what we deserve, i've got another trashy thing to add.

Everything everyone is coming up with is worth something, only a supernatural 
act can stop man as a whole from developing an a.i. on the level of a quite a 
successful virus that can tackle its environment quite well.     But?  why now? 
Why not happen ages ago?   I think something from the super-natural has always 
been happening to stop man from "over-tackling" his environment with self 
reduplicating instant menial work machines,   so if I suddenly disappear,   you 
can probably guess god handled me, if you were to think like me, just like how 
he handled everyone else that thinks simply enough to actually get it to happen.

It all disappears,   if you over-achieve, god will sort you out.   And god 
prefers dummies that don't hit super civilizational efficiency, cause they dont 
cause waves that make the whole thing not work at all.



--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te3b75f99e52289f7-M72448678ee4b82e54d432db4
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[WriterOfMinds]

John, now that you've elaborated some more, I think you're just trying to say: 
there may exist an algorithm that performs lossy compression in the general 
case, but achieves lossless compression on certain specialized datasets. As far 
as I know, that's possible. It doesn't create some middle ground between lossy 
and lossless compression, though. It does one or the other, depending on what 
data you feed in. By definition, lossless = not lossy. The compressed result 
either contains all of the information that was in the original, or does not 
contain all of the information that was in the original.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-Mb33cc2ac78413baac5e5026f
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: group on Telegram if you would like please join here

[korrelan]

I joined a few days ago and left/ deleted it yesterday, it's going to take a 
month of inactivity for my account to be fully removed apparently.  

Typical chat environment, no considered posts/ content... to much posturing/ 
nonsense and not enough thinking.

I read all the various feeds... er... no.

:)

https://sites.google.com/view/korrtecx 

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tc7f6ce83ae20b45a-M7528f8796e4ed52614d1a3cf
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: group on Telegram if you would like please join here

[rouncer81]

not into installs thanks.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tc7f6ce83ae20b45a-M91491eb5c47d180eea408b14
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Missing Data

[immortal . discoveries]

The lossless compressed result is not the same file anymore, it doesn't contain 
all the information therefore. It just so happens to re-generate back the exact 
file. Some lossless decompressors take longer to decompress also. And Lossy 
(when expect all data to return) takes very long - maybe years before someone 
reinvents your missing data. It's there, all the information of the original 
results are thee, it really is, it just takes longer.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M052c7fda97e482534001bb7c
Delivery options: https://agi.topicbox.com/groups/agi/subscription