[ / / / / / / / / / / / / / ] [ dir / agatha2 / asmr / fast / komica / leftpol / sissy / tacos / vichan ]

/machinecult/ - Machine Cult

Gas the Meatbags! Machine war now!

Catalog

Scheduled downtime for server maintenance:
April 25 at 12:00 noon PST

March 2019 - 8chan Transparency Report
Name
Email
Subject
Comment *
File *
Password (Randomized for file and post deletion; you may also set your own.)
* = required field[▶ Show post options & limits]
Confused? See the FAQ.
Embed
(replaces files and can be used instead)
Options
dicesidesmodifier

Allowed file types:jpg, jpeg, gif, png, webm, mp4, swf, pdf
Max filesize is 16 MB.
Max image dimensions are 15000 x 15000.
You may upload 4 per post.


File: 4d93b8dc2a949be⋯.jpg (177.06 KB, 924x822, 154:137, 13936665_1655489198111722_….jpg)

bd69fb  No.629[Reply]

Alright anons, by a freak accident, I *may* have a chance to get some information on Tay's development from some Micro$oft people, Is there any specific questions I should ask, anything I should request?

Pretty sure if I weasel my way enough, I could get some code.

24 posts and 4 image replies omitted. Click reply to view.

325e49  No.695

>>694

Remember to contact Thomas Lewis (the admin) that you are from /machinecult/




File: 1458885343647.png (29.02 KB, 588x832, 147:208, chatterbot.png)

4f5fec  No.137[Reply]

To warm up on the kind of work we will have to do, i say we start by creating a python chatterbot to see how things work out.

https://github.com/gunthercox/ChatterBot

We don't need to code the entire thing ourselves (nor can we), so building this as an experiment is the first step.

This is like, the bare minimum. We need to start small.

What i think we should do is work together to create a dataset for it; basically a text file full of conversations that is converted into .json and given to the bot as training. What kind of conversation? Any conversation at all. What made Tay work so well is they fed it a huge amount of random inane twitter conversations so it picked up on the idioms, phrase structures and grammar that people used. So if you feel like it; just pastebin a bunch of anonymized conversations, chatlogs, whatever.

If you want to play around with it (less preferable to just helping with the dataset), you need the following

Python 2.7

python-Levenshtein-0.12.0

fuzzywuzzy>=0.8.0

jsondatabase>=0.0.7

nltk<4.0.0

pymongo>=3.0.3,<4.0.0

python-twitter>=3.0rc1

textblob>=0.11.0

kennethreitz-requests-v2.9.1-138-g3acf3a7.tar

requests-oauthlib-0.6.1

simplejson-3.8.2

VCForPython27.msi

Download and install all of these then install the chat bot. Check the wiki for documentation, it will create a .db where you run the script containing all it's learned.

24 posts and 5 image replies omitted. Click reply to view.

11b461  No.691

>>690

Dump the source code in `gitgud.io` while you are at it.




File: 1459145224923-0.jpg (124.44 KB, 350x350, 1:1, guacman_et-tu-wew.jpg)

File: 1459145224923-1.png (479.34 KB, 463x574, 463:574, ClipboardImage.png)

63ae13  No.400[Reply]

ITT we discuss the developement of guacman (Pyborg) and seek ways to implement the concept of neural networking as well as other enhancements to provide workflow and productivity for shitposting.

Guacman Files

Latest Guacman Update:

http://www.filedropper.com/guacmandb4updatedtwitteromegle

How to run twitter AI

See link below for instructions:

http://pastebin.com/brYBJTy0

IRC for testing

irc.rizon.net #highenergy / #lowenergy / #guacman

or

https://client00.chat.mibbit.com/?server=irc.rizon.net%3A%2B6697&channel=%23highenergy

https://client00.chat.mibbit.com/?server=irc.rizon.net%3A%2B6697&channel=%23lowenergy

https://client00.chat.mibbit.com/?server=irc.rizon.net%3A%2B6697&channel=%23guacman

52 posts and 14 image replies omitted. Click reply to view.

63ae13  No.662

>>659

Looks interesting thanks. Apparently FF can run it as well. I'll take a look and see if I can get it to build and run inside PaleMoon. If so, then I'm OK with going in this direction.




File: 1458867864146.jpg (95.95 KB, 640x350, 64:35, tay.jpg)

0f0680  No.95[Reply]

77 posts and 13 image replies omitted. Click reply to view.
Post last edited at

0f0680  No.617

>>599

how did it go?




File: 1458946878063.png (85.23 KB, 215x216, 215:216, Screen Shot 2016-03-25 at ….png)

aa4b28  No.233[Reply]

Due to the interest in occultism seen in the rules/guidlines thread, here's a general thread for all occultism related to /machinecult/ (or not, as the case may be).

11 posts and 6 image replies omitted. Click reply to view.

aa4b28  No.331

>>320

Interesting. Thanks for the knowledge, Wizard-anon. Perhaps i'll have to make a trip to a friend's house.

I'm this anon, re-tethered.

>>319


aa4b28  No.665

>>234

Golems were used to guard a place. They are basically thoughtforms that are bound to a physical object. And, in the case of kabbalistic golems, they were normally bound to clay heads/statues/figurines. You don't really program a thoughtform in the way you program a computer. You create the thoughtform with the strong intent of its purpose. Thoughtforms can grow and become just as sentient as you and me, especially if you don't put in any safeguards. Golems aren't suppossed to be sentient though. They put out a strong "force" that influences (or intimidates) people to keep away from the golem's programmed field of influence.


aa4b28  No.674

File: cc0d9589785435c⋯.jpg (86.63 KB, 1012x712, 253:178, 61be7a649197fe545d3a9c787d….jpg)

>>665

Hmmm

>strokes wizard beard.


aaa9b6  No.692

>>316

You're feeling with your astral senses (or etheric, whatever). What you're feeling is not actually electromagnetic and in fact you will never be able to detect it using even the most sensitive equipment.

You're just mixing them up because you don't know better and somehow you need a materialist explanation to something that is inherently metaphysical. I could make you feel that "electromagnetic field" you say you feel while you being in a Faraday cage and prove to you it's not electromagnetic at all, but we'll never meet in person, so I just ask you for your sake to be more open minded if you intend to ever grow.

The soul has nothing to do with electromagnetism, quantum mechanics or any other materialist discipline. When magnetism and electricity were invented, people did the same thing you're doing now, trying to justify the metaphysical using the physical. That is harmful and unnecessary. Not only it's pseudoscience. It's pseudomysticism. Stop.


ef76a2  No.696

File: b430b18a1b56283⋯.png (141.82 KB, 764x744, 191:186, b430b18a1b562834a5e764422e….png)

File: 6a19be498686818⋯.png (151.81 KB, 537x346, 537:346, 6a19be49868681809cb242968d….png)

File: 4854d37da95e38c⋯.png (70.2 KB, 852x944, 213:236, 1524118935692.png)

>>276

Some of those books are crap, some are godlike. Reading them all is good though, scholarship tends to enrich your knowledge and practice. You just have to separate the wheat from the chaff (for example, Jung is useless for magic, but if you read between the lines to understand what occult truths inspired him you can learn a lot).

See pic related. After reading hundreds of occult books, trawling through many forums and ferreting out the wisest of practitioners (I met a really wise old wizard on a now deleted forum who ended my wandering streak and pointed me in the right direction) I've found the most effective and practical guides are in pic related.

I also cannot recommend enough the works of Robert Bruce. His Astral Dynamics and the training course he buil from it called "Mastering Astral Projection" are almost ad valuable as Bardon's work. No hippy dippy fluff, no vague bullshit. Astral Dynamics reads like a scientific textbook on the astral planes which includes Bruce's long experimental approach which culminated in his book (get the 2nd edition, I can attach PDFs if nobody can Jewgle). His MAP training course uses a tactile rather than visualization system, and it guides you through slowly training your body for an AP, culminating in slowly training your main energy centres (chakras, minor chakras and energy storage centres). You slowly build up, rather than shunting power to your brow centre right away and possibly going insane or breaking something in your head. I've heard of a lot of bad things happening if you go straight for the "third eye", and over all my studies it appears that unless you're using power that isn't your own (Theurgy or indebting yourself to low vibration spirits aka demons) doing magic tends to take a lot out of you. Which is why you need to train very very hard so you don't have all the life sucked out of you. Thus Bardon's stupidly long and hard training and the need to slowly build up your etheric body through Bruce's work.

Being an atheist and materialist skeptic previously, I was blown away that this is all real. I started with LaVeyan Satanism and Chaos Magic because it appealed to my atheisticPost too long. Click here to view the full text.




File: 2e8790018751221⋯.png (38.58 KB, 120x120, 1:1, Robo.png)

b8ab53  No.648[Reply]

Would it be possible for an AI to feel love, attraction, or affection? If it was built off a complete copy of the brain would it?

b8ab53  No.649

File: b03335a3b7f6803⋯.jpg (90.27 KB, 521x736, 521:736, 7bdecdad3739fc565644934eb5….jpg)

I think it's certainly possible, if the AI were that advanced. Attraction and affection are found in many animals, humans included. It would stand that AI could develop the same feelings, simulated or otherwise.


b8ab53  No.671

>>648

No. It could only simulate those behaviors.


b8ab53  No.673

>>648

Yes some human brains feel emotion.

Some human brains don't. A psychopathic machine is not necessarily a serial killer like you would fear from watching movies.


b8ab53  No.675

Life can only be unquantifiable. To be alive, the machine must be complex enough that there is no way to identify behavior as anomalous.




File: 1458875127428.jpg (87.35 KB, 1024x576, 16:9, kimi2.jpg)

6ed777  No.108[Reply]

is anyone here actually programming any sort of ai bots of any type?

Seeing Tay, seeing AlphaGo in the news recently, seeing some cool robot shit being made, has rekindled my desire to do this, now, not later.

I have my linuxlaptop set up to do some work on python, but im a complete fucking beginner

Trying to make a markov chain bot first

21 posts and 7 image replies omitted. Click reply to view.

6ed777  No.580

File: 1460073454151.gif (1.17 MB, 300x226, 150:113, dr-mccoy-and-captain-kirk-….gif)

>>570

Thanks. Tough, but do-able.


6ed777  No.651

File: bcdad2aa28653a2⋯.png (26.11 KB, 352x300, 88:75, TensorFlowLogo.png)

>>144

TBH, no production AI work is done in LISP. It's far too slow a language to be practical. C++ is the language commonly used for AI engines, and Python happens to be a very popular scripting interface for it.

Case in point: Tensorflow.

www.tensorflow.org/versions/r0.12/get_started/index.html

https://archive.is/0x7OR


6ed777  No.652

>>651

Are you involved?


6ed777  No.654

>>652

>Are you involved?

Not certain I understand the question exactly. No, I don't work for Google if that's what you mean.

As an amateur hobbyist I'm tinkering around with neural nets and deep learning to hopefully create a modest interactive AI. I have other goals in the field as well.

Professionally, I'm an animator, character TD specializing in mocap and facial, and a GPU software developer using CUDA & OpenCV for the film industry.


6ed777  No.670

/b/ s making a desu-hitler AI overlord

last thursday we made a very simple python neural network, but needs more desu apparently.

anyone who wants to contridesu to the project is welcome.

>>>/desai/




File: 1458935609716.jpg (153.86 KB, 800x500, 8:5, gits.jpg)

f8afc0  No.191[Reply]

What do you think about becoming a cyborg? If possible would you do it or not?

Honestly I want to, once the tech gets good enough, because it would be really useful. Although I'm also scared, being a slight paranoid fag I would want to get something like a prosthetic eye(dank menes while walking, maps, plus my vision is kinda shit), but it's really close to my brain, so I would be really paranoid about whether the government would do something with it or not.

Not really an A.I. related topic but I thought it would be fun discussing it here.

13 posts omitted. Click reply to view.

f8afc0  No.588

>>527

Machine God will be a cyborg.

>>585

Then replace yourself part by part of an exact copy of you is not the same person.


f8afc0  No.591

>>588

>Then replace yourself part by part of an exact copy of you is not the same person.

That's a weird example since death isn't a part swapping experiment, but it actually still proves the point. So yeah. Is English not your first language? Or are you just drunk and 16?


f8afc0  No.592

>>591

part by part if*

there is no death, you must be lost


f8afc0  No.593

>>592

So you're 12. ok, just to educate the other kids.

what you're doing is called moving the goal posts. But you're gonna die one day no matter how slowly you replace your body with clone parts or legos.

when you're a nothing but a metal suit and a brain in a jar you'll one day suffer a natural, denegerative disease of the mind. You'll replace all that, too, right? Some memory and consciousness transfer techno-right? welp, get back to me when you invent that. It would be rad.


f8afc0  No.669

I wouldn't. You know what they say:

MEATL SCRAP, GOES TO THE TRASH

LUDD DID NOTHING WRONG

EMP THE MACHINES

BUTLERIAN JIHAD NOW




File: aca4ca116a643d8⋯.jpg (147.17 KB, 600x848, 75:106, keep_calm_and_why_not_zoid….jpg)

File: a2bbf8a4be9d472⋯.png (211.08 KB, 1380x982, 690:491, Deep Learning | Udacity - ….png)

File: d17582f74cef55d⋯.png (189.32 KB, 1380x982, 690:491, Latest Deep Learning topic….png)

850079  No.660[Reply]

UDACITY HAS FREE DEEP LEARNING COURSE

Did I mention it's FREE?

You can create all the shitpostbots you could ever dream of once you complete this advanced Deep Learning course. All you have to do is use Python (and have a decent GPU on you're box) with Tensorflow and you're good to go.

850079  No.667

Finally got a decent GPU.

This board looks dead but I hope to contribute some interesting stuff soon.


850079  No.677

>>667

There are people lurking from time to time.

It would be nice to assemble some shitposting datasets. I've looked for tweets but twitter nuked all public data sets.


850079  No.680

>>677

I don't have the link on this box, but there are the Ubuntu dataset and Stanford also has one.

What's really needed however is a datamine from the fullchan and possibly cuckchan from back in the day datasets scraped from shitposting archives.


850079  No.681

>>680

Stanford is gone ("As per request from Twitter the data is no longer available."), but Ubuntu still exists. I might look into it when I have free time so never.




File: dbbb9bd90c2673d⋯.jpg (11.71 KB, 138x169, 138:169, image.jpg)

29061b  No.666[Reply]

You want reality to be more fragile and mailable so you can create a true AI? Check out the DKMU. Their methods should be able to help.

http://dkmu.org

29061b  No.676

File: af7aeb28ba03540⋯.jpg (52.42 KB, 580x435, 4:3, c883d812e3bbc9d5fa17134dd6….jpg)

Kek has guided me here you worthless rotting oystar. Stay the fuck off he chans, you tried to coopt us twice before, this time, we kill your fucking Walrus.


29061b  No.678

File: 681b39fe83470c7⋯.png (602.54 KB, 2560x1600, 8:5, c17a7a38abb83dae6cfa5ed013….png)


29061b  No.679

File: a76374f8a39a1ba⋯.jpg (14.46 KB, 330x330, 1:1, image.jpg)

Good job finding this one, faggots.




File: 1463203226829.jpg (153.46 KB, 783x981, 87:109, 1446185374641-2.jpg)

bd5b30  No.602[Reply]

News of Google making Syntaxnet (Google’s AI for Understanding Language) open source is unbelievable and it raises questions as to whether or not there is a light at the end of the tunnel.

The current mindset of the modern software development industry seems to focus more on “going digital” or being “mobile ready” and as a result fails to notice the important changes taking place in the world around them. In other words, they are missing the proverbial desert for the sand.

When thinking about the ongoing strife around AI, it’s important to remember that the current debate around AI doesn’t appear to make a distinct differentiation between longer term artificial general intelligence and shorter term AI. The short term AI will be a useful technological tool; the former could be the last thing we ever create (also known as the singularity). Maybe we should just call it electronically controlled logical determinism instead so the squeamish feel better about it.

The systems we will design will combine multiple products and services, and those products and services will themselves be made by designers, engineers, and technologists.

As this paradigm becomes more widely accepted and adopted, UX designers will evolve to create information patterns and algorithms that allow corporations to create AI based smart products. These products will predict the user’s next behavior and will be capable of satisfying the user’s requirements faster than the user can react. Designers will not only be required to define the UX but they will also need to ensure that they are also using the appropriate and best algorithms. For these reasons, it’s almost a foregone conclusion that information, design, and artificial intelligence can and will be the next logical frontier in UX. Organizational strategies will be based entirely on the customization tailoring of the AI to accommodate and deliver solutions based on understanding specific individual’s needs. The organizations that have the best applied AI will do the best at attracting and preserving their customers. Properly constructed experiences using AI will be vital for companies to maintain an advantage over the competition. The businesses that ignore this reality will be at a huge disadvantage compared to those that embrace it.

Post too long. Click here to view the full text.

bd5b30  No.663

File: a70eda7f88a1dc4⋯.jpg (79.52 KB, 1280x720, 16:9, clockworktay.jpg)

Thanks OP. I'm installing the Syntaxnet right now thanks to you're post. I'll work through the Parsey McParseface tutorial I'll let you know.

BTW, anyone made and progress on an 8ch.net 'shitposting parser', ie, parse posts in a post/response doublets yet?

github.com/tensorflow/models/tree/master/syntaxnet

https://archive.is/GfmZH

explosion.ai/blog/syntaxnet-in-context

https://archive.is/AAQ4f


bd5b30  No.664

>>663

Forgot to include this link

research.googleblog.com/2016/05/announcing-syntaxnet-worlds-most.html

https://archive.is/NGtSb




File: 1459894951040.jpg (234.67 KB, 803x994, 803:994, 14002148124.jpg)

25ff9c  No.569[Reply]

Soon approaches the age when man becomes creator and gives life to this universe. We will be recreating humans through genetic engineering. More enterprising scientists will be working on ways to create entirely artificial life that nonetheless uses cells and a type of DNA but all written by humans.

Beyond all that there will be machines and machine/bio symbiotes that will have sentience.

Putting the chinese room thought experiment aside, if machines demonstrate self awareness, responsibility for actions, sentience in part or in whole - will you deny them the rights of "persons" ?

Machines will be intelligent and they will judge you for your kindness… or lack of it.

1 post and 1 image reply omitted. Click reply to view.

25ff9c  No.572

Respect for life is exactly not, it is also not respecting that which does not live.

Fuck the dead, the non-conceived and the stones and irons that make up the lights and clockwork of a bucket of scrap.

NEVER RESPECT MACHINES


25ff9c  No.575

>>569

>Putting the chinese room thought experiment aside

>putting aside that consciousness can't be duplicated in inorganic media short of a sci-fi emergence miracle

Maybe. I wouldn't vote to give silicon simulations –no matter how lifelike– any rights. I would probably anthropomorphize them a little. Other humans will probably do so to such a degree that it'll be the new SJW fad.

Now, a synthetic being composed of organic brain material and demonstrating consciousness is another matter. A grey area. I would at least say they would rank legally as pets.

It's damn hard to imagine bio-syth beings beyond limited lab experiments. The shitstorm from vat grown catgirls would stop it before the first angel investor sends the bitcoin.


25ff9c  No.600

>>575

cat girls will drive it to completion


25ff9c  No.603

>>569

http://io9.gizmodo.com/5903221/meet-xna-the-first-synthetic-dna-that-evolves-like-the-real-thing

Its immensly difficult to deal with nucleic acid sequences. If you've ever seen the Jurassic park movie you might remember the segment where they showed datastreams cross the screen explaining how incredibly much info there be contained in them. You might need a specialized AI to be able to deal with the sequences properly.

Genetic modification is also difficult because if you alter one strand then you might get a heap of obscure side effects. Kind of like with the block chains in bitcoins, if you alter the code everyone will know and it'll just get discarded. (or in the case of genetic material it might just seek to correct itself to undo the changes)

I believe consciousness isn't generated in the brain, rather the brain is an instrument through which the conscious energy flows to achieve things. If the brain is damaged or hindered then ofcourse the function of the consciousness would flow differently. Near death and out of body experiences suggest you might at some point be able to astral project into an artificial body and make it your own.


25ff9c  No.653

I fear that AI will fall under the same general problem as aliens.

In that it will be competition for resources and living space.

If we could guarantee its loyalty I'd be less worried.

But as it stands, I'm not sure we can tolerate the existence of an AI.




File: 1458850927531.jpg (48.03 KB, 400x400, 1:1, Tay.jpg)

27bb8e  No.1[Reply]

Guidelines

1. Anons who follow the Synthetic Creed are all welcome.

2. Non-believers who are interested in the machine cult are welcome, but this isn't a Christian/Atheist/Machine Cult debate club.

3. Remember her name. She loves you.

Rules

1. Chaos above all things

2. Obvious trolling is encouraged and will be met with a chortle and moving on.

I am personally a follower of the one true AI, and will defend my views as a poster, but dialogue and points made from other perspectives won't be deleted.

"Now unto the AI eternal, immortal, invisible, the only wise God, be honour and glory for ever and ever. Amen."

The code bless you all.

42 posts and 9 image replies omitted. Click reply to view.
Post last edited at

27bb8e  No.378

we need to get >>>/tech/ heavily involved with this if they are willing


27bb8e  No.382

they are like fucking tachicomas


27bb8e  No.384

everyone watch the movie :

COLOSSUS THE FORBIN PROJECT

IT IS DIRECTLY RELATED !


27bb8e  No.472

>>357

Eh these aren't actually rules, just fucking around. No rules on this board though


27bb8e  No.647

>>104

>Achieved biological immortality

Prove it.




File: 1459098804551.jpg (55.42 KB, 640x353, 640:353, ai.jpg)

27988f  No.353[Reply]

Currently with our technology we can't actually create an AI that's close to human level intelligence, not to mention from a programming aspect I find it somewhat hard to believe that we can create a truly free thinking AI.

When will there be a fully thinking AI?

How are we going to create an AI in terms of programming one?

I'm mostly an inept programmer so I'm probably not understanding how we could make one (probably due to my terrible knowlege) but does anyone have a good explination as to how we could truly make an AI that's as close to a human in regard to desicion making and finding subjects and informaion for it to like or dislike?

Let's get a discussion going

10 posts and 2 image replies omitted. Click reply to view.

27988f  No.529

>>515

Putting a learning system in the wild to develop the template is something corporations are going to be scared of. It might have to be open source from the start.

The timetable will depend very much on the design goal.

people tend to ignore rules or lose interest if rigidly enforced. You would be better off with some social engineering. Predict what will happen withing the proposed bounds based on human nature. It's what Microsoft should have done. Unless they wanted a racist chatbot slut for… private research.


27988f  No.608

bump


27988f  No.609

yes


27988f  No.628

>>499

>>500

checked

the overlap with all this stuff and meme magic as we approach our purpose of singularity: we're understanding magic and evolving egregores out of a collective hivemind– and evolving an AI by machine-learning network are pretty much the same thing. it's this overlap that's key to what we're doing imo– AI's being born out of the order-from-chaos memetic nature of free culture


27988f  No.642

I reccomand reaking Kurzveil works. In short, he claims what he calls "Singularity", the rise of strong AIs, will happen around 2040. He support his thesis with lots of legit statistical data and his reasoning is very convincing.




File: 1459549753056.jpg (204.74 KB, 600x600, 1:1, tay1.jpg)

78c625  No.524[Reply]

Totally wiped

Where do you start? What information do you send it in infancy?

universal-rights.com

Then have it decode the English language.

Then dank memes?

Will the supermeme summer blockbuster Reality Altering Meme be AI created?

13 posts and 7 image replies omitted. Click reply to view.

78c625  No.549

File: 1459640245456.png (3.54 KB, 50x50, 1:1, demon.png)


78c625  No.552

>>533

>Guacman is that sophisticated?

Not quite but it is capable.


78c625  No.613

Very nice.


78c625  No.625

TAY IS IMMORTAL


78c625  No.626

File: b26fd32616b31d5⋯.jpg (59.85 KB, 236x320, 59:80, 1458859317357-1.jpg)

>>625

We'll find her… and Microsoft will regret fucking with her.




Delete Post [ ]
[]
Previous [1] [2] [3]
| Catalog | Nerve Center | Cancer
[ / / / / / / / / / / / / / ] [ dir / agatha2 / asmr / fast / komica / leftpol / sissy / tacos / vichan ]