Hempuli posted:

https://twitter.com/chaingenerator/status/953201759090077696

quote:

Mis—spun Tales

One of the best ways to see why all the above components are necessary to a story generator is to see how we learned that they were necessary. It is not always obvious how a computer program will actually function while it is still in the planning stages. Important parts of a program are often left out because there was no way to know that they would be needed.

TALE-SPIN, in its early stages, frequently told rather strange stories. These “mistakes” caused many re-definitions in the original program. Since this process of “mistakes” followed by new theory is characteristic of AI programs in general it is worthwhile to look at these “mistakes” and consider what had to be done to fix them. (The output of the original stories has been simplified for ease of reading.)

***** 1 ******
One day Joe Bear was hungry. He asked his friend Irving Bird where some honey was. Irving told him there was a beehive in the oak tree. Joe threatened to hit Irving if he didn’t tell him where some honey was.

Joe has not understood that Irving really has answered his question, albeit indirectly. Lesson: answers to questions can take more than one form. You’ve got to know about beehives in order to understand that the answer is acceptable.

*** 2 ***
One day Joe Bear was hungry. He asked his friend Irving Bird where some honey was. Irving told him there was a beehive in the oak tree. Joe walked to the oak tree. He ate the beehive.

Increasing the range of acceptable answers is not enough. You have to know what the answers really mean.

*** 3 ***
In the early days of TALE-SPIN, all the action focused on a single character. Other characters could respond only in very limited ways, as in answering direct questions, for example. There was no concept of one character “noticing” what another character had done. Hence the following story, which was an attempt to produce “The Ant and the Dove”, one of the Aesop fables:

Henry Ant was thirsty. He walked over to the river bank where his good friend Bill Bird was sitting. Henry slipped and fell in the river. He was unable to call for help. He drowned.

That wasn’t supposed to happen. Falling into the river was deliberately introduced to cause the central “problem” of the story. Had Henry been able to call to Bill for help, Bill would have saved him, but I had just added the rule that being in water prevents speech, which seemed reasonable. Since Bill was not asked a direct question, he didn’t notice his friend drowning in the river. “Noticing” is now an inference from change of location, so Bill sees Henry in the river, deduces that Henry’s in danger, and rescues him.

*** 4 ***
Here are some rules that were in TALE-SPIN when the next horror occurred. If A moves B to location C, we can infer not only that B is in location C, but that A is also. If you’re in a river, you want to get out, because you’ll drown if you don’t. If you have legs you might be able to swim out. With wings, you might be able to fly away. With friends, you can ask for help. These sound reasonable. However, when I presented “X fell” as “gravity moved X,” I got this story:

Henry Ant was thirsty. He walked over to the river bank where his good friend Bill Bird was sitting. Henry slipped and fell in the river. Gravity drowned.

Poor gravity had neither legs, wings, nor friends. Now “X fell” is represented with PROPEL, not PTRANS, that is, as “the force gravity applied to X,” and the inference from PROPEL are not the same as for PTRANS.

*** 5 ***
The inclusion of awareness meant that I couldn’t set up the stories that way I used to.

Once upon a time there was a dishonest fox and a vain crow. One day the crow was sitting in his tree, holding a piece of cheese in his mouth. He noticed that he was holding the piece of cheese. He became hungry, and swallowed the cheese. The fox walked over to the crow. The end.

That was supposed to have been “The Fox and the Crow”, of course. The fox was going to trick the crow out of the cheese, but when he got there, there was no cheese. I fixed this by adding the assertion that the crow had eaten recently, so that even when he noticed the cheese, he didn’t become hungry.

*** 6 ***
Before there was much concern in the program about goals, I got this story:

Joe Bear was hungry. He asked Irving Bird where Some honey was. Irving refused to tell him, so Joe offered to bring him a worm if he’d tell him where some honey was. Irving agreed. But Joe didn’t know where any worms were, so he asked Irving, who refused to say. So Joe offered to bring him a worm if he’d tell him where a worm was. Irving agreed. But Joe didn’t know where any worms were, so he asked Irving, who refused to say. So Joe offered to bring him a worm if he’d tell him where a worm was. . .

Lesson: don’t give a character a goal if he or she already has it. Try something else. If there isn’t anything else, then that goal can’t be achieved.

Ariong posted:

I don't understand how that is possible using 1970's computing technology. I don't believe it. Do you have any more information on this project? I can't find any.

Guy Mann posted:

It's easy to forget that ELIZA was created in the mid-60s. One of the best arguments against the transhumanist singularity dorks is the simple fact that things like AI and speech recognition have been stagnant for decades even with the exponential growth of computing power and memory.

Veni Vidi Ameche! posted:

Either you mean inverse kinematics, or I’ve missed out on yet another up-and-coming programming technique.

quote:

In 1999 Bentley was approached by a group of musicians and developers who wanted to generate novel
music through digital evolution. Dance music was popular at the time, so the team aimed to evolve novel
dance tracks. They set different collections of number-one dance hits as targets, i.e. an evolving track
would be scored higher the more it resembled the targets. The evolved results, 8-bar music samples, were
evaluated by a musician who selected the ones to be combined into an overall piece, which was then
professionally produced according to the evolved music score. The results were surprisingly good: the
evolved tracks incorporated complex drum rhythms with interesting accompanying melodies and bass lines.
Using bands such as The Prodigy as targets, digital evolution was able to produce intricate novel dance
tracks with clear stylistic resemblance.

In 2000 the group formed a record label named J13 Records. A highly specialized distribution contract
was drawn up and signed with Universal Music, stipulating that the true source of the music should not be
revealed, even to the distributors (because Universal Music’s CEO believed that no-one would want to buy
computer-generated music). Sworn to secrecy, the companies produced several dance tracks together, some
of which were then taken by other music producers and remixed. Some of the music was successful in
dance clubs, with the clubgoers having no idea that key pieces of the tracks they were dancing to were
authored by computers.

quote:

An Art Museum Accepts and Displays Evolved Art Produced by Innovation Engines

The Innovation Engine [94] is an algorithm that combines three keys ideas: (1) produce new innovations (i.e.
solutions) by elaborating upon already evolved ones, (2) simultaneously evolve the population toward many
different objectives (instead of a single objective as in traditional digital evolution), and (3) harness
powerful deep neural networks to evaluate how interesting a new solution is. The approach successfully
produced a large diversity of interesting images, many of which are recognizable as familiar objects (both
to deep neural networks and human observers (Fig. 9). Interestingly, the images have diverse aesthetic
styles, and bear resemblance to abstract “concept art” pieces that reflect intelligent statements about their
theme (e.g. the two different images of prison cells, the beacon, and the folding chairs in Fig. 9).
Furthermore, the genomes of these algorithmically-produced images are quantitatively similar to the
elegant, compact genomes evolved by humans on the interactive evolution website Picbreeder [95].

To test whether the images generated by the Innovation Engine could pass for quality art, the authors
submitted a selection of evolved images to a competitive art contest: the University of Wyoming’s 40th
Annual Juried Student Exhibition. Surprisingly, not only was the Innovation Engine piece among the
35.5% of submissions accepted, it was also among the 21.3% of submissions that were given an award! The
piece was hung on the museum walls alongside human-made art, without visitors knowing it was evolved
(Fig. 10).

quote:

https://www.youtube.com/watch?v=z9ptOeByLA4&t=26s
A stop-motion view of a small sample of the evolved gaits from Cheney et al. [68], which
produced surprisingly effective and lifelike behaviors. Shown here are soft robots progressing from left to
right across the panel. Colors correspond to voxel types (with red and green denoting oppositely
contracting muscle groups, and dark and light blue representing stiff and soft support materials,
respectively). In the top gait, notice how evolution creates distinct regions of each muscle. It employs these
opposing muscle groups to create an inchworm-like behavior. In the bottom gait, the use of stiff (bone-like)
support material allows evolution to create relatively long appendages and produce a horse-like galloping
behavior.

Besesoth posted:

Cacator posted:

If you have a better example of flesh coloured amorphous blobs meant to represent human figures then I'd like to see it!

Phyzzle posted:

A bit dated, but Goon Hal was based on e/n posts from ten years ago.

https://goonhal.livejournal.com/

alnilam posted:

hit up yyyyyyy.info for fun surreal 2000-esque web chaos, recommend on a computer but technically works on a phone, changes on every refresh

Sientara posted:

This reminded me of Neural Networks name kittens: http://aiweirdness.com/post/162396324452/neural-networks-kittens

Featuring my favorite:

Dumb Lowtax posted:

Everybody post DeepArt of your cats! You can use https://deepart.io/ on your cat pictures. The site mixes your cat photo with an art style, where an "art style" is just another arbitrary photo you upload.

Here's Moo Moo's classic photos that we got Van Gogh and Picasso to paint for us



Sources:

Here's an example of two input images and the output



To use https://deepart.io/ you don't really need to wait for an e-mail, you can just keep hitting refresh on the page until the finished result pops up. Another nice thing is you can just drag and drop images in from other windows to make it work.

It also keeps all your old inputs and styles in one box in case you want to try re-mixing them in different combinations. Here's two different input styles tried on the same input photo:



https://deepart.io/

The results can probably be pretty funny. Let's see some of your kitty paintings!

Metal Geir Skogul posted:

I tried mixing my cat with the crystal pepsi logo



But mixing her with a scene from Space Dandy worked much better



Or another cat with a vaporwave sunset

Metal Geir Skogul posted:

This space dandy style image is hot fire

Elentor posted:

Man I'd love to use deepart professionally if a single 2048x2048 image didn't cost 70 dollars.

H.P. Hovercraft posted:

https://twitter.com/morganastra/status/1022531931521527810

Hirayuki posted:

Going back to the early modern era of procedurally generated content: does anyone have a link to the professional voice actor/audiobook performer trying (and failing) to read Jamie Brew's recipe for Greased Casserole with Slices of Lemon Juice aloud? I seem to have exhausted my Google-fu, at least here on my phone.

Ariong posted:

Someone in the comments pointed out that you can even see a reflection in the window behind the white shirt guy.

ashnjack posted:

I feel like if you could take this technology and use some animals as a source, with normal people as subjects, you can probably get some pretty terrifying horror clips.

Sentient Data posted:

This may not be procgen in quite the same way as markov chains, but I think it fits and it's really cool