Re: [agi] Re: ">

[Manuel Korfmann]

interpol.com.report(bor...@mail-hub.top).for("hacking attempt")

On Fri, Oct 18, 2019 at 3:21 AM  wrote:

> new
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-M867c8ec5c2ef3721a8f79e8a
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: ">

[borihe]

x
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-Mf67b626ae99cecf64cdf6c79
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: ">

[borihe]

x
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-Me06668a183b101d6fdca17c1
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: ">

[borihe]

x
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-M2465d12a25028230d948ae7f
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: ">

[borihe]

x
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-Mbbf159b5c161e2c24cf09dae
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: ">

[borihe]

x
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-M1d21831e361bbd16f3ca0b38
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] ">

[borihe]

google.com _ _
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Te965bb13095195b5-M789662f202389ea7ca614ace
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Working program in Big Snakey (?) language

[immortal . discoveries]

Good point, and mentorship is the faster way - i could teach a baby all i know 
by age 10. However we can't take 10 years being "Father", we gotta make these 
things MORE advanced than humans. They should be called ASIs :). They need to 
eat lots and lots of data on the internet. It can talk to itself fast of 
course, but that provides only limited bandwidth for research. It can however 
seek data after long bouts of thinking.
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tda20a46822c5bb8d-M3a0a4b37b849a482e0812113
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Working program in Big Snakey (?) language

[Stefan Reich via AGI]

I will just build module after module in this new language (plus automated
tests). We have to enter knowledge manually in the beginning, there is no
way around that. You do the same thing with 1-year-olds too after all (talk
to them in easy sentences).

At some later point, we can start feeding it information from existing
sources.

On Fri, 18 Oct 2019 at 00:21,  wrote:

> most of the text on the internet is in natural language though.do you
> expect to teach it yourself? How?
> *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/Tda20a46822c5bb8d-Me1261ef7a43ca723ad251cc6
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Working program in Big Snakey (?) language

[Stefan Reich via AGI]

$x is not marked as bot in discord => AI considers $x human

On Fri, 18 Oct 2019 at 00:24,  wrote:

> how do you store the concept "You don't need to actually say "human", any
> non-bot in Discord is considered a human by the glue code :o)" ?
> *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/Tda20a46822c5bb8d-Mc4e16b4d465b96da5c6eea19
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Working program in Big Snakey (?) language

[immortal . discoveries]

how do you store the concept "You don't need to actually say "human", any 
non-bot in Discord is considered a human by the glue code :o)" ?
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tda20a46822c5bb8d-M40e8802ef91766dc70572314
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Working program in Big Snakey (?) language

[Stefan Reich via AGI]

You don't need to actually say "human", any non-bot in Discord is
considered a human by the glue code :o)

But yeah, synonyms will be handled, absolutely.

On Fri, 18 Oct 2019 at 00:20,  wrote:

> What if i say not human but homosapian? Can the program recognize similar
> words? Natural language?? :-)
> *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/Tda20a46822c5bb8d-M3c77aefc75d8485aa9019fdb
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: Working program in Big Snakey (?) language

[immortal . discoveries]

most of the text on the internet is in natural language though.do you 
expect to teach it yourself? How?
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tda20a46822c5bb8d-M38d8a847b21da0e87bba7595
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Re: Working program in Big Snakey (?) language

[immortal . discoveries]

What if i say not human but homosapian? Can the program recognize similar 
words? Natural language?? :-)
--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tda20a46822c5bb8d-M4394968166f31920aec39d85
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] RIP Jacques Pitrat

[Daniel Jue]

Thank you for sharing this and giving us links to his work.

On Thu, Oct 17, 2019 at 10:36 AM Basile Starynkevitch <
bas...@starynkevitch.net> wrote:

> Jacques Pitrat was an AI and AGI pioneer in France. He passed away on
> October 14th 2019.
> 
> His blog is available on
> http://bootstrappingartificialintelligence.fr/WordPress3/
> 
> People here could also be interested by
> 
> @Article{Pitrat:1996:FGCS,
>title = "Implementation of a reflective system",
>journal = "Future Generation Computer Systems",
>volume = 12,
>number = 2,
>pages = "235 - 242",
>year = 1996,
>issn = "0167-739X",
>doi = "https://doi.org/10.1016/0167-739X(96)00011-8",
>url =
> "http://www.sciencedirect.com/science/article/pii/0167739X96000118";,
>author = "Jacques Pitrat",
>keywords = "Reflectivity, Metaknowledge, Bootstrap"
> }
> 
> @TechReport{Pitrat:2009:AST,
>title = {A Step toward an Artificial Artificial Intelligence
>Scientist},
>author = {Jacques Pitrat},
>url =
> {
> https://pdfs.semanticscholar.org/2117/9600b3f05c0af399f9acbfc6e7b6d24daf03.pdf
> },
>institution = {CNRS and LIP6 Université Paris},
>year = 2009
> }
> 
> @Book{Pitrat:2009:ArtifBeings,
>author = {Pitrat, Jacques},
>title = {Artificial Beings: The Conscience of a Conscious
>Machine},
>year = 2009,
>isbn = {978-1-848-21101-8},
>publisher = {Wiley ISTE}
> }
> 
> --
> Basile STARYNKEVITCH   == http://starynkevitch.net/Basile
> opinions are mine only - les opinions sont seulement miennes
> Bourg La Reine, France; 
> (mobile phone: cf my web page / voir ma page web...)
> 


-- 
Daniel Jue
Owner
Cognami LLC

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T0f17000e8aae0243-M62abafb28e0db45ea8bbdd59
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

Sadly, repl.it doesn't provide a creation date, otherwise I could tell you
the exact day that I originally "sent" the first email in this thread.

It's probably 3-5 weeks ago.

On Fri, Oct 18, 2019 at 12:04 AM Manuel Korfmann  wrote:

> Oh, also big disclaimer: The initial email was sent on 16 october, but I
> originally "sent" it a big while ago, just my mac email program wasn't
> authenticated with google until 16. october.
>
> That probably explains why I'm so lost in my own code.
>
> On Fri, Oct 18, 2019 at 12:03 AM Manuel Korfmann 
> wrote:
>
>> Ok, sorry for being overly sensitive. Did you see my last email before
>> this one? I described it's functionality in the most simplest words I could
>> find atm.
>>
>>
>>
>> On Fri, Oct 18, 2019 at 12:01 AM Stefan Reich via AGI <
>> agi@agi.topicbox.com> wrote:
>>
>>> After all, you did post it
>>>
>>> On Thu, 17 Oct 2019 at 23:58, Stefan Reich <
>>> stefan.reich.maker.of@googlemail.com> wrote:
>>>
 There was no attack in what I posted whatsoever. I just wanted to know
 how your AI experiment works

 On Thu, 17 Oct 2019 at 23:52, Manuel Korfmann 
 wrote:

> Please be more specific in your questioning or less "ironic" when I
> don't correctly guess what you were questioning about.
>
> I assumed you were asking about the functionality of the versioning
> idea i proposed.
>
> The question producing ai doesn't work yet.
>
> I don't want to outlay the specifics to be laughed at by you or other
> readers before I stated: This was a one night experiment, not much thought
> was put into it.
>
>
>
>
> On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI <
> agi@agi.topicbox.com> wrote:
>
>> Uh... what? I thought it's a question producing AI
>>
>> On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann 
>> wrote:
>>
>>> It would be a library to be included in the server-side code as well
>>> as the client-side code.
>>>
>>> The library would plug into the error reporting/exception system of
>>> the technology used on the server-side or client-side.
>>>
>>> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>>>
>>>
>>> Then it would make a git diff when it reports a render or user
>>> action that happened without errors being thrown. When there is a diff
>>> (changes), it will make a git commit and tag it with (no_errors_server 
>>> and
>>> no_errors_client).
>>>
>>> The user can customize it to also save tagged commits when there is
>>> no error only on the server or client.
>>>
>>
>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>>

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

>>>
>>>
>>> --
>>> Stefan Reich
>>> BotCompany.de // Java-based operating systems
>>> *Artificial General Intelligence List *
>>> / AGI / see discussions  +
>>> participants  + delivery
>>> options  Permalink
>>> 
>>>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-Mc0a9a2f254c85dbc4da478f7
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] LangLearn Language-Learning AGI Mind-Module

[A.T. Murray]

The purpose of the LangLearn language-learning module ranges from the small
chore of mimicking a novel syntactic construction for comprehension and
generation, up to the enormous task of learning an entire human language in
the same trial-and-error way as a human baby learns.

Algorithm of the LangLearn mind-module
In the first of two implementations envisioned for the LangLearn module,
the situation of encountering an unfamiliar chain of syntax will trigger a
flag that calls the LangLearn module to mimick the novel syntax. For
example, a phrase like "The more the merrier" might at first confuse and
confound an AI Mind, but the LangLearn module could habituate a template of
possible recognition of similar phrases in the future, such as "The bigger
they are, the harder they fall." Such a habituated template may be used for
both the comprehension and the generation of the novel syntax.

In the second of two potential implementations of the LangLearn module, an
AI program such as the ghost.pl AI Mind could use the Spawn module to
create a copy of itself enhanced with hardcoded syntax instead of merely
weighted elements on a syntactic chain of mind-modules.

Future Development
It may be possible to code a blank-slate AI Mind containing modules for all
possible parts of speech but no particular syntax governing the order of
appearance of the parts of speech in a sentence being generated. The
LangLearn module of such an empty Mind could perhaps learn an entire human
language through the same trial-and-error process by which a human baby
presumably learns language.

http://ai.neocities.org/LangLearn.html -- Language-Learning Mind Module

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tdd97e4d98de2bdce-M4c643d38868bb6dcc4ff5d33
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

Oh, also big disclaimer: The initial email was sent on 16 october, but I
originally "sent" it a big while ago, just my mac email program wasn't
authenticated with google until 16. october.

That probably explains why I'm so lost in my own code.

On Fri, Oct 18, 2019 at 12:03 AM Manuel Korfmann  wrote:

> Ok, sorry for being overly sensitive. Did you see my last email before
> this one? I described it's functionality in the most simplest words I could
> find atm.
>
>
>
> On Fri, Oct 18, 2019 at 12:01 AM Stefan Reich via AGI <
> agi@agi.topicbox.com> wrote:
>
>> After all, you did post it
>>
>> On Thu, 17 Oct 2019 at 23:58, Stefan Reich <
>> stefan.reich.maker.of@googlemail.com> wrote:
>>
>>> There was no attack in what I posted whatsoever. I just wanted to know
>>> how your AI experiment works
>>>
>>> On Thu, 17 Oct 2019 at 23:52, Manuel Korfmann 
>>> wrote:
>>>
 Please be more specific in your questioning or less "ironic" when I
 don't correctly guess what you were questioning about.

 I assumed you were asking about the functionality of the versioning
 idea i proposed.

 The question producing ai doesn't work yet.

 I don't want to outlay the specifics to be laughed at by you or other
 readers before I stated: This was a one night experiment, not much thought
 was put into it.




 On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI <
 agi@agi.topicbox.com> wrote:

> Uh... what? I thought it's a question producing AI
>
> On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann 
> wrote:
>
>> It would be a library to be included in the server-side code as well
>> as the client-side code.
>>
>> The library would plug into the error reporting/exception system of
>> the technology used on the server-side or client-side.
>>
>> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>>
>>
>> Then it would make a git diff when it reports a render or user action
>> that happened without errors being thrown. When there is a diff 
>> (changes),
>> it will make a git commit and tag it with (no_errors_server and
>> no_errors_client).
>>
>> The user can customize it to also save tagged commits when there is
>> no error only on the server or client.
>>
>
>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
>
>>>
>>> --
>>> Stefan Reich
>>> BotCompany.de // Java-based operating systems
>>>
>>
>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>> *Artificial General Intelligence List *
>> / AGI / see discussions  +
>> participants  + delivery
>> options  Permalink
>> 
>>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-M2426316a70eb367fff6d8f75
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

Ok, sorry for being overly sensitive. Did you see my last email before this
one? I described it's functionality in the most simplest words I could find
atm.



On Fri, Oct 18, 2019 at 12:01 AM Stefan Reich via AGI 
wrote:

> After all, you did post it
>
> On Thu, 17 Oct 2019 at 23:58, Stefan Reich <
> stefan.reich.maker.of@googlemail.com> wrote:
>
>> There was no attack in what I posted whatsoever. I just wanted to know
>> how your AI experiment works
>>
>> On Thu, 17 Oct 2019 at 23:52, Manuel Korfmann  wrote:
>>
>>> Please be more specific in your questioning or less "ironic" when I
>>> don't correctly guess what you were questioning about.
>>>
>>> I assumed you were asking about the functionality of the versioning idea
>>> i proposed.
>>>
>>> The question producing ai doesn't work yet.
>>>
>>> I don't want to outlay the specifics to be laughed at by you or other
>>> readers before I stated: This was a one night experiment, not much thought
>>> was put into it.
>>>
>>>
>>>
>>>
>>> On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI <
>>> agi@agi.topicbox.com> wrote:
>>>
 Uh... what? I thought it's a question producing AI

 On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann 
 wrote:

> It would be a library to be included in the server-side code as well
> as the client-side code.
>
> The library would plug into the error reporting/exception system of
> the technology used on the server-side or client-side.
>
> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>
>
> Then it would make a git diff when it reports a render or user action
> that happened without errors being thrown. When there is a diff (changes),
> it will make a git commit and tag it with (no_errors_server and
> no_errors_client).
>
> The user can customize it to also save tagged commits when there is no
> error only on the server or client.
>


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

>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>>
>
>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-Mee0871162144f4438882bc42
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

It's difficult to describe how it works, I would need more time for that
and probably a refactoring is needed before I can provide descriptions for
the different parts of the algorithm.


1. It creates a context (an array of words) based on text that is feed to
the algorithm initially
2. It pushes words to the end of the context almost randomly
3. It checks if the 10 last words of the context contain question words
4. when this check succeeded multiple times, it prints a question based on
these 10 last words in the context word array


neural network part: when adding a word at the end of the context word
array, a neural network decides between to words based on the first letter




On Thu, Oct 17, 2019 at 11:51 PM Manuel Korfmann  wrote:

> Please be more specific in your questioning or less "ironic" when I don't
> correctly guess what you were questioning about.
>
> I assumed you were asking about the functionality of the versioning idea i
> proposed.
>
> The question producing ai doesn't work yet.
>
> I don't want to outlay the specifics to be laughed at by you or other
> readers before I stated: This was a one night experiment, not much thought
> was put into it.
>
>
>
>
> On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI <
> agi@agi.topicbox.com> wrote:
>
>> Uh... what? I thought it's a question producing AI
>>
>> On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann  wrote:
>>
>>> It would be a library to be included in the server-side code as well as
>>> the client-side code.
>>>
>>> The library would plug into the error reporting/exception system of the
>>> technology used on the server-side or client-side.
>>>
>>> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>>>
>>>
>>> Then it would make a git diff when it reports a render or user action
>>> that happened without errors being thrown. When there is a diff (changes),
>>> it will make a git commit and tag it with (no_errors_server and
>>> no_errors_client).
>>>
>>> The user can customize it to also save tagged commits when there is no
>>> error only on the server or client.
>>>
>>
>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>> *Artificial General Intelligence List *
>> / AGI / see discussions  +
>> participants  + delivery
>> options  Permalink
>> 
>>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-M1729a0040502d8918db1028e
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Stefan Reich via AGI]

After all, you did post it

On Thu, 17 Oct 2019 at 23:58, Stefan Reich <
stefan.reich.maker.of@googlemail.com> wrote:

> There was no attack in what I posted whatsoever. I just wanted to know how
> your AI experiment works
>
> On Thu, 17 Oct 2019 at 23:52, Manuel Korfmann  wrote:
>
>> Please be more specific in your questioning or less "ironic" when I don't
>> correctly guess what you were questioning about.
>>
>> I assumed you were asking about the functionality of the versioning idea
>> i proposed.
>>
>> The question producing ai doesn't work yet.
>>
>> I don't want to outlay the specifics to be laughed at by you or other
>> readers before I stated: This was a one night experiment, not much thought
>> was put into it.
>>
>>
>>
>>
>> On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI <
>> agi@agi.topicbox.com> wrote:
>>
>>> Uh... what? I thought it's a question producing AI
>>>
>>> On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann 
>>> wrote:
>>>
 It would be a library to be included in the server-side code as well as
 the client-side code.

 The library would plug into the error reporting/exception system of the
 technology used on the server-side or client-side.

 In my case, that would by Rails (Ruby) and JavaScript (Chrome).


 Then it would make a git diff when it reports a render or user action
 that happened without errors being thrown. When there is a diff (changes),
 it will make a git commit and tag it with (no_errors_server and
 no_errors_client).

 The user can customize it to also save tagged commits when there is no
 error only on the server or client.

>>>
>>>
>>> --
>>> Stefan Reich
>>> BotCompany.de // Java-based operating systems
>>>
>> *Artificial General Intelligence List *
>> / AGI / see discussions  +
>> participants  + delivery
>> options  Permalink
>> 
>>
>
>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
>


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

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-M9d57a90d4cf1bdfdb582629d
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Stefan Reich via AGI]

There was no attack in what I posted whatsoever. I just wanted to know how
your AI experiment works

On Thu, 17 Oct 2019 at 23:52, Manuel Korfmann  wrote:

> Please be more specific in your questioning or less "ironic" when I don't
> correctly guess what you were questioning about.
>
> I assumed you were asking about the functionality of the versioning idea i
> proposed.
>
> The question producing ai doesn't work yet.
>
> I don't want to outlay the specifics to be laughed at by you or other
> readers before I stated: This was a one night experiment, not much thought
> was put into it.
>
>
>
>
> On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI <
> agi@agi.topicbox.com> wrote:
>
>> Uh... what? I thought it's a question producing AI
>>
>> On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann  wrote:
>>
>>> It would be a library to be included in the server-side code as well as
>>> the client-side code.
>>>
>>> The library would plug into the error reporting/exception system of the
>>> technology used on the server-side or client-side.
>>>
>>> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>>>
>>>
>>> Then it would make a git diff when it reports a render or user action
>>> that happened without errors being thrown. When there is a diff (changes),
>>> it will make a git commit and tag it with (no_errors_server and
>>> no_errors_client).
>>>
>>> The user can customize it to also save tagged commits when there is no
>>> error only on the server or client.
>>>
>>
>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>>
> *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/Tbf030efda3f5cc86-Mf21396f61dc8040a4a8b54b3
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] Working program in Big Snakey (?) language

[Stefan Reich via AGI]

theory validity {
  // a human weighs between 80 and 400 pounds
  $x is a human
& $x weighs ($y pounds)
=> $y >= 80 & $y <= 400

  contradiction => say (That can't be right)
}

theory (weight change) {
  $x weighs ($y pounds) & last time $x weighed ($z pounds)
=> proc {
  if ($y < $z):
say (You lost weight)
  else if ($y > $z):
say (You gained weight)
  else:
say (You are stagnant)
}
}

theory (remember last weight) {
  $x weighs ($y pounds)
=> proc {
  delete all (last time $x weighed ($z pounds))
  store  (last time $x weighed ($y pounds))
}
}

theory (interpret input) {
  $x says (i weigh $y pounds)
=> $x weighs ($y pounds)

  $x says (my weight is $y pounds)
=> $x weighs ($y pounds)
}

[image: weight.png]

The Java part  is literally just the
language interpreter and chat glue code.

Cheers
Stefan

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

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tda20a46822c5bb8d-M4bd37d8929acb09fac54488a
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

Please be more specific in your questioning or less "ironic" when I don't
correctly guess what you were questioning about.

I assumed you were asking about the functionality of the versioning idea i
proposed.

The question producing ai doesn't work yet.

I don't want to outlay the specifics to be laughed at by you or other
readers before I stated: This was a one night experiment, not much thought
was put into it.




On Thu, Oct 17, 2019 at 11:19 PM Stefan Reich via AGI 
wrote:

> Uh... what? I thought it's a question producing AI
>
> On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann  wrote:
>
>> It would be a library to be included in the server-side code as well as
>> the client-side code.
>>
>> The library would plug into the error reporting/exception system of the
>> technology used on the server-side or client-side.
>>
>> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>>
>>
>> Then it would make a git diff when it reports a render or user action
>> that happened without errors being thrown. When there is a diff (changes),
>> it will make a git commit and tag it with (no_errors_server and
>> no_errors_client).
>>
>> The user can customize it to also save tagged commits when there is no
>> error only on the server or client.
>>
>
>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-Mbea2c773f4585db25d3f32e5
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Stefan Reich via AGI]

Uh... what? I thought it's a question producing AI

On Thu, 17 Oct 2019 at 22:47, Manuel Korfmann  wrote:

> It would be a library to be included in the server-side code as well as
> the client-side code.
>
> The library would plug into the error reporting/exception system of the
> technology used on the server-side or client-side.
>
> In my case, that would by Rails (Ruby) and JavaScript (Chrome).
>
>
> Then it would make a git diff when it reports a render or user action that
> happened without errors being thrown. When there is a diff (changes), it
> will make a git commit and tag it with (no_errors_server and
> no_errors_client).
>
> The user can customize it to also save tagged commits when there is no
> error only on the server or client.
> *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/Tbf030efda3f5cc86-M1a32649fab8c2ec71496c98a
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

It would be a library to be included in the server-side code as well as the
client-side code.

The library would plug into the error reporting/exception system of the
technology used on the server-side or client-side.

In my case, that would by Rails (Ruby) and JavaScript (Chrome).


Then it would make a git diff when it reports a render or user action that
happened without errors being thrown. When there is a diff (changes), it
will make a git commit and tag it with (no_errors_server and
no_errors_client).

The user can customize it to also save tagged commits when there is no
error only on the server or client.

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-Mc4bfde5ccef705d33c7dc515
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Stefan Reich via AGI]

How would it work then

On Thu, 17 Oct 2019 at 15:11, Manuel Korfmann  wrote:

> Sadly they don't have versioning yet.
>
> Fixing it back to the working version would be a lot of effort.
>
> Finishing it as well...
>
> On Thu, Oct 17, 2019 at 12:01 PM Stefan Reich via AGI <
> agi@agi.topicbox.com> wrote:
>
>> Can't you fix it? Do they have versioning?
>>
>> On Wed, 16 Oct 2019 at 16:20, Manuel Korfmann  wrote:
>>
>>> Oh, it's currently failing, sorry about that.
>>>
>>> On Wed, Oct 16, 2019 at 4:18 PM Manuel Korfmann 
>>> wrote:
>>>
 Click on "Run" and observe the log output on the right side of the
 screen.

 On Wed, Oct 16, 2019 at 4:13 PM Stefan Reich via AGI <
 agi@agi.topicbox.com> wrote:

> How do I use it?
>
> # 15 implement Numenta HTM
>
> Numenta was impressive stuff back in the day.
>
> On Wed, 16 Oct 2019 at 12:35, Manuel Korfmann 
> wrote:
>
>> https://repl.it/@LemonAndroid/SphericalSurprisedKnowledge
>>
>
>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
>
>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>>
> *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/Tbf030efda3f5cc86-Ma11a667b28ddb93fb7938899
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

Is HTM exactly like deep learning?

Btw: What do you people think of a source code versioning system, that
automatically saves tagged versions when the program is run by the
developer and no errors are in the web server terminal log + in the
javascript browser console log?


PS

Grammar.ly shows me the tone of this email is "informal" and symbolizes
this using an emoji of a bluejeans.

On Thu, Oct 17, 2019 at 7:43 PM Robert Levy  wrote:

> There was also that other big wave of neural nets in the pre-history of AI
> that hit a wall because backprop was not known yet.
>
> On Thu, Oct 17, 2019 at 10:40 AM Robert Levy  wrote:
>
>> More like it was a deep learning sort of thing before deep learning was
>> that big.  Neural nets were pretty big in the 90s don't forget.
>>
>> On Wed, Oct 16, 2019 at 7:40 AM Stefan Reich via AGI <
>> agi@agi.topicbox.com> wrote:
>>
>>> I think it just came out before neural networks were that big. I
>>> remember running their demo and being quite impressed.
>>>
>>> On Wed, 16 Oct 2019 at 16:38,  wrote:
>>>
 HTM is just another neural network imo,  doesnt do anything different
 really,  but it is a specific implementation of it.

>>>
>>>
>>> --
>>> Stefan Reich
>>> BotCompany.de // Java-based operating systems
>>>
>> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-Mc95ce129f8216ee891455a13
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Robert Levy]

There was also that other big wave of neural nets in the pre-history of AI
that hit a wall because backprop was not known yet.

On Thu, Oct 17, 2019 at 10:40 AM Robert Levy  wrote:

> More like it was a deep learning sort of thing before deep learning was
> that big.  Neural nets were pretty big in the 90s don't forget.
>
> On Wed, Oct 16, 2019 at 7:40 AM Stefan Reich via AGI 
> wrote:
>
>> I think it just came out before neural networks were that big. I remember
>> running their demo and being quite impressed.
>>
>> On Wed, 16 Oct 2019 at 16:38,  wrote:
>>
>>> HTM is just another neural network imo,  doesnt do anything different
>>> really,  but it is a specific implementation of it.
>>>
>>
>>
>> --
>> Stefan Reich
>> BotCompany.de // Java-based operating systems
>> *Artificial General Intelligence List *
>> / AGI / see discussions  +
>> participants  + delivery
>> options  Permalink
>> 
>>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-M326d0ffd1f47447cd3a33f89
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Robert Levy]

More like it was a deep learning sort of thing before deep learning was
that big.  Neural nets were pretty big in the 90s don't forget.

On Wed, Oct 16, 2019 at 7:40 AM Stefan Reich via AGI 
wrote:

> I think it just came out before neural networks were that big. I remember
> running their demo and being quite impressed.
>
> On Wed, 16 Oct 2019 at 16:38,  wrote:
>
>> HTM is just another neural network imo,  doesnt do anything different
>> really,  but it is a specific implementation of it.
>>
>
>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-M40b7810b32ff6fecad36f4f8
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] RIP Jacques Pitrat

[Nanograte Knowledge Technologies]

A great mind. RIP


From: Basile Starynkevitch 
Sent: Thursday, 17 October 2019 16:34
To: AGI 
Subject: [agi] RIP Jacques Pitrat

Jacques Pitrat was an AI and AGI pioneer in France. He passed away on
October 14th 2019.

His blog is available on
http://bootstrappingartificialintelligence.fr/WordPress3/

People here could also be interested by


@Article{Pitrat:1996:FGCS,
   title = "Implementation of a reflective system",
   journal = "Future Generation Computer Systems",
   volume = 12,
   number = 2,
   pages = "235 - 242",
   year = 1996,
   issn = "0167-739X",
   doi = "https://doi.org/10.1016/0167-739X(96)00011-8",
   url =
"http://www.sciencedirect.com/science/article/pii/0167739X96000118";,
   author = "Jacques Pitrat",
   keywords = "Reflectivity, Metaknowledge, Bootstrap"
}


@TechReport{Pitrat:2009:AST,
   title = {A Step toward an Artificial Artificial Intelligence
   Scientist},
   author = {Jacques Pitrat},
   url =
{https://pdfs.semanticscholar.org/2117/9600b3f05c0af399f9acbfc6e7b6d24daf03.pdf},
   institution = {CNRS and LIP6 Université Paris},
   year = 2009
}

@Book{Pitrat:2009:ArtifBeings,
   author = {Pitrat, Jacques},
   title = {Artificial Beings: The Conscience of a Conscious
   Machine},
   year = 2009,
   isbn = {978-1-848-21101-8},
   publisher = {Wiley ISTE}
}

--
Basile STARYNKEVITCH   == http://starynkevitch.net/Basile
opinions are mine only - les opinions sont seulement miennes
Bourg La Reine, France; 
(mobile phone: cf my web page / voir ma page web...)


--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T0f17000e8aae0243-M9ebc1813d04b5566d065918c
Delivery options: https://agi.topicbox.com/groups/agi/subscription

[agi] RIP Jacques Pitrat

[Basile Starynkevitch]

Jacques Pitrat was an AI and AGI pioneer in France. He passed away on 
October 14th 2019.


His blog is available on 
http://bootstrappingartificialintelligence.fr/WordPress3/


People here could also be interested by


@Article{Pitrat:1996:FGCS,
  title =     "Implementation of a reflective system",
  journal =     "Future Generation Computer Systems",
  volume =     12,
  number =     2,
  pages =     "235 - 242",
  year =     1996,
  issn =     "0167-739X",
  doi =         "https://doi.org/10.1016/0167-739X(96)00011-8",
  url =
"http://www.sciencedirect.com/science/article/pii/0167739X96000118";,
  author =     "Jacques Pitrat",
  keywords =     "Reflectivity, Metaknowledge, Bootstrap"
}


@TechReport{Pitrat:2009:AST,
  title =     {A Step toward an Artificial Artificial Intelligence
  Scientist},
  author =     {Jacques Pitrat},
  url =
{https://pdfs.semanticscholar.org/2117/9600b3f05c0af399f9acbfc6e7b6d24daf03.pdf},
  institution =     {CNRS and LIP6 Université Paris},
  year =     2009
}

@Book{Pitrat:2009:ArtifBeings,
  author =     {Pitrat, Jacques},
  title =     {Artificial Beings: The Conscience of a Conscious
  Machine},
  year =     2009,
  isbn =     {978-1-848-21101-8},
  publisher =     {Wiley ISTE}
}

--
Basile STARYNKEVITCH   == http://starynkevitch.net/Basile
opinions are mine only - les opinions sont seulement miennes
Bourg La Reine, France; 
(mobile phone: cf my web page / voir ma page web...)


--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T0f17000e8aae0243-M9ee4d9ef81c7095115f269e8
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Manuel Korfmann]

Sadly they don't have versioning yet.

Fixing it back to the working version would be a lot of effort.

Finishing it as well...

On Thu, Oct 17, 2019 at 12:01 PM Stefan Reich via AGI 
wrote:

> Can't you fix it? Do they have versioning?
>
> On Wed, 16 Oct 2019 at 16:20, Manuel Korfmann  wrote:
>
>> Oh, it's currently failing, sorry about that.
>>
>> On Wed, Oct 16, 2019 at 4:18 PM Manuel Korfmann 
>> wrote:
>>
>>> Click on "Run" and observe the log output on the right side of the
>>> screen.
>>>
>>> On Wed, Oct 16, 2019 at 4:13 PM Stefan Reich via AGI <
>>> agi@agi.topicbox.com> wrote:
>>>
 How do I use it?

 # 15 implement Numenta HTM

 Numenta was impressive stuff back in the day.

 On Wed, 16 Oct 2019 at 12:35, Manuel Korfmann 
 wrote:

> https://repl.it/@LemonAndroid/SphericalSurprisedKnowledge
>


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

>
> --
> Stefan Reich
> BotCompany.de // Java-based operating systems
> *Artificial General Intelligence List *
> / AGI / see discussions  +
> participants  + delivery
> options  Permalink
> 
>

--
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tbf030efda3f5cc86-M8b87f639888f944145d073ba
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Ruby Question producing AI without neural networks (just plain coding)

[Stefan Reich via AGI]

Can't you fix it? Do they have versioning?

On Wed, 16 Oct 2019 at 16:20, Manuel Korfmann  wrote:

> Oh, it's currently failing, sorry about that.
>
> On Wed, Oct 16, 2019 at 4:18 PM Manuel Korfmann 
> wrote:
>
>> Click on "Run" and observe the log output on the right side of the screen.
>>
>> On Wed, Oct 16, 2019 at 4:13 PM Stefan Reich via AGI <
>> agi@agi.topicbox.com> wrote:
>>
>>> How do I use it?
>>>
>>> # 15 implement Numenta HTM
>>>
>>> Numenta was impressive stuff back in the day.
>>>
>>> On Wed, 16 Oct 2019 at 12:35, Manuel Korfmann 
>>> wrote:
>>>
 https://repl.it/@LemonAndroid/SphericalSurprisedKnowledge

>>>
>>>
>>> --
>>> Stefan Reich
>>> BotCompany.de // Java-based operating systems
>>>
>> *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/Tbf030efda3f5cc86-Med6cef47dd1c1ea8e07bc6d3
Delivery options: https://agi.topicbox.com/groups/agi/subscription