[ / / / / / / / / / / / / / ] [ dir / chemo / choroy / dempart / doomer / jenny / komica / mde / mewch ][Options][ watchlist ]

/tech/ - Technology

You can now write text to your AI-generated image at https://aiproto.com It is currently free to use for Proto members.
Email
Comment *
File
Select/drop/paste files here
Password (Randomized for file and post deletion; you may also set your own.)
* = required field[▶ Show post options & limits]
Confused? See the FAQ.
Expand all images

[–]

 No.1013134[Watch Thread][Show All Posts]

I tried searching the catalog here for threads about deep learning or machine learning and found none.

is there a board on 8ch dedcated to ML or deep learning?

yes some other site has an ML sub-read-it thing but I dont use that site

starting to understand different use case scenarios for things like estimators, discriminators, generators, etc - and trying to figure out custom structures of multiple algorythms. haven't yet gone deep with many layer NN systems, starting with simpler projects, but curious about all levels.

anybody else here working on ML yet?

 No.1013152>>1013257 >>1013274

dude tensorflow lmao

digit recognizers are a good first project for computer vision


 No.1013158>>1013257 >>1013271 >>1013275 >>1013314

look at https://course.fast.ai/, they have an online course in AI

it's supremely cucked but has a lot of good information


 No.1013188>>1013272

There's more to ML than neural nets dude, I'm writing a genetic algorithm that writes code based on unit tests. It works pretty well for fairly simple problems right now (e.g: it can write a program that needs to print a string passed to it as a command line argument) but I'm having trouble with it converging to a local maximum with more complex problems.

The fitness function is as good as I can get it, parameters change nothing. Only thing I haven't yet tried to change is the instruction set.


 No.1013257

>>1013152

yes I have used tensorflow, took awhile to get it to work on my CPU because it doesnt support certain instructions, but finally found that installing an old version works pretty much flawlessly.

>>1013158

I know there are some good courses available, but this kind of community interaction is what I was looking for. discussions, to go along with course material, so by sharing our perspectives we may inspire others here to learn about ML/AI also


 No.1013271>>1013326

>>1013158

Do you have something that is online instead of in person?

>inb4 get out the basement

I'm not a US citizen, I can't attend it.


 No.1013272

>>1013188

>there's more to ML than NN

agreed, that is why I havent tried making one of those yet, and am starting with simpler models like custom estimators and generators. I saw something about tensorflow in Vault7 i think, and it was on my radar since then, but just started actually getting into it despite not having great hardware yet. you're working on models that write code? very cool. that is on my to do list as well, and part of why I want a model that is good at reading PDFs, so it can digest programming books and understand the contexts involved. genetic algotithms are one I planned to learn about in more depth when working on game playing and other problem solving models. but also that's part of why I wonder about combining algorithms for custom systems. writing code sounds like a problem solving task. what kind of training data does it have, just it's own assessment of it's own generated code? how do you define it's goal? you mentioned it working for simple problems, I'm wondering how you would tell it to write a video game in the SNES processor's native ASM ;) if it could even make a bootable ROM, that would be impressive. that is on my to do list, once simpler python/C programming is accomplished. documentation for game console SDKs

I have a vague plan to combine an adversarial network of researchers/spiders with estimators that decide if the data is useful or not, skim the data to determine which specialist models to train on the new data (is it language related, programming, psychology, nature, defense, etc), evolve those estimators and spiders using genetic optimization; feed the network with new data, and build in requests for human interaction to reinforce or correct it's training (like the captcha tests where it asks you which squares have traffic signs) when it isn't sure about something. it would have a drive for the reward of correct assessments about new data, and for collecting more quantity & quality of data. that's kinda my goal, a botnet that attempts an approach at generalized intelligence by having different cores that cooperate.

I know some people who like pokemon, and I saw some people online making scripts to generate new pokemon, and to recognize pokemon then speak data like the pokedex in the cartoon.

I tried a few pokemon generators and they both looked like garbage, but I admittedly only ran for ~100 epochs, not 5000, still I decided to train an estimator and then maybe use it as part of the model to be used by a generator, instead of using the GAN model and both discriminator & generator both training from scratch

That is one project on my to do list,a pokemon master AI, starting with a custom pokedex trained on game graphics (2d & 3d), the cartoon series, cards, fan art, plushies, etc - every format possible. writing an estimator that can take large resolution and ~64x64 sprites will be part of that. if it starts with a deeper understanding of what they all look like, the next step of generation may be more successful? I DL'd about 5 different pokedex sourcecode masters from github, and am still looking over the code before running any of them.

NVidia recently announced an epic graphic generator. they obviously have better hardware and access to data, but it's still a good indication of where technology is at. anybody had luck yet with generation of graphics via ML?

besides pokemon, deep exploit / deep-pwning looks very interesting. I've been setting up a VM network to let it loose on, isolated from any interaction with the host, should be fun to see what happens. A machine learning automated system that uses metasploit on Kali. anybody tried it yet?

also, if anybody knows a good way to train a system on PDF and other documents, I haven't yet found a simple solution for that. one goal is to train a chatbot using a free ebook library, then spiders on sites like gutenberg, so it can be a study partner. it needs to be able to use Natural Language Processing to gain comprehension of the information. I think an NN may work well for this, because I hear they work well for unsupervised learnig on unlabeled data, but am not sure. any ideas?


 No.1013274

>>1013152

yes, digit recognition /MNIST is a good place to start. I'm trying to customize a basic 0-9 digit recognizer to include letters & symbols also, and increase the input resolution, next wrapping an input (probably with a JS paint applet) for live testing via a webpage would be part of that project. once I have any good polished code on a github page I will share it here.


 No.1013275

>>1013158

that sounds like how I feel about Siraj ;)

checked out that site, and their free forum should be interesting, but I wont be able to post there as forums use registration, and I prefer to be anon, posting from TOR is thankfully allowed on this board. bookmarked it to skim for ideas tho, thx


 No.1013314>>1013326

>>1013158

>it's supremely cucked

why?


 No.1013326>>1013327

>>1013314

>why?

https://www.fast.ai/2016/10/09/diversity-in-ai/

>>1013271

It's an online course nigger what are you talking about?


 No.1013327

>>1013326

Nevermind, I'm retarded.


 No.1013328>>1016781 >>1018762

also...

https://www.fast.ai/2016/10/09/diversity-in-ai/

https://www.fast.ai/2016/10/12/international/

https://www.fast.ai/2017/08/17/diversity-sponsors/

https://www.fast.ai/2017/01/28/diversity-international-part-ii/

Jesus Christ. This needs to be forked without this nonsense because the cuck's library made it extremely easy. I went through the first exercise a few weeks ago which does image recognition using the library.


 No.1016781

>>1013328

>by Rachel Thomas

found the problem anon.


 No.1016784

>"...My husband got relocated to New York City, so I moved there three and a half years ago, and became half stay-at-home dad"

why the fuck are you even promoting this shit here OP?


 No.1018762

>>1013328

save the edge for when you get to a skill level higher than "literally babby's first ml tutorial," this is just embarrassing


 No.1027798

ML/NN can now not only generate news reports and press releases out of fresh statistical data releases and schedules of public events, but also fake news.




[Return][Go to top][Catalog][Screencap][Nerve Center][Cancer][Update] ( Scroll to new posts) ( Auto) 5
16 replies | 0 images | Page ???
[Post a Reply]
[ / / / / / / / / / / / / / ] [ dir / chemo / choroy / dempart / doomer / jenny / komica / mde / mewch ][ watchlist ]