/robowaifu/ - DIY Robot Wives

Advancing robotics to a point where anime catgrill meidos in tiny miniskirts are a reality.

Reports of my death have been greatly overestimiste.

Still trying to get done with some IRL work, but should be able to update some stuff soon.

#WEALWAYSWIN

Max message length: 6144

Drag files to upload or
click here to select them

Maximum 5 files / Maximum size: 20.00 MB

Captcha
no cookies?
More

(used to delete files and postings)


Open file (32.03 KB 400x400 FXCY9fGv.jpg)
THE LANGUAGE PROBLEM Robowaifu Technician 11/20/2020 (Fri) 13:22:39 No.6937 [Reply]
'Sup anons? I am here to remind you guys something important, TO DO YOUR RESEARCH IN MULTIPLE LANGUAGES. Our mutual language is English however that is not enough. We need people who can speak those 3 important languages: Japanese, Chinese, Russian. I've been learning japanese for 4 months and with the help of dictionary I am able to understand basic stuff. Here is the point, there is a whole another world out there. 1) Chinese: Chinese people work under such hard circumstances and put really much effort into their jobs. Nearly none of the projects are translated into English since Google is banned at China. However there are a lot of great stuff there, I mean like even Microsoft runs their virtual woman project there. Since it is too hard for me to learn I generally use DeepL (best translator out there) and Baidu (Chinese search engine) and read latest researches and projects there. I wish I knew chinese well, in that case I would be able to find not-so-popular webpages and grab more information on topics. 2) Japanese: Even though good Japanese projects gets translated into English most of the research there gets translated only when the projects is ready to publish and sometimes they are too hard to find. I try to read as much AI papers in Japanese as I can. Scientists there do great stuff. I've seen a lot of robowomen projects there. You can also find some 3d printing projects for anime girls. Really worth looking. 3) Russian: Russian is the least important one in my opinion. But a professor of mine graduated from a university there and he has a lot of academic books that aren't translated to English. You would be amazed at how much work they have on stuff such as algorithm theory, artificial intelligence, computer science. Most of the stuff there are focused on "science" part of CS. So they are theory-weighted. So right now we need people who can speek Japanese and Chinese (Korean would be good as well, but there isn't that much of research there tbh) Using DeepL is enough to understand most of the pages but only a person with fluent Chinese/Japanese would be able to find goldmines deep there. I am pretty sure that there are hundreds of Chinese people working on robowaifu related projects that we are not aware of. Same applies for Japanese people, but since Chinese people are in a much worse situation it becomes really hard to find those people. Anyone has some recommendations? I wish I had time and skills to learn all those languages but I can only afford to learn one and I am going with Japanese since I have a dream of moving there in the future. We need to brainstorm on this issue.
7 posts omitted.
Anon, I've been keeping tabs on Japanese projects on Twitter, the robowaifu guys there love to retweet anything interesting. I notice most of them use Raspberry Pi, Nvidia Jetson Nano, M5Stack, and the multitudes of Shenzhen arduino clones and microcontrollers. What they mainly got different is they got their own seemingly Japan-exclusive servos or for some reason are able to afford Dynamixels. As for language processing, if we are training chatbots they will be dependent on the specific language unless they just all make cute noises like Chii -- well our mascot is Chii so a cute noise generator waifu will be more universally appealing without the need for translations.
>>6975 We miss you no longer reporting on updates from those guys. That really added some interesting flavor and inspiration to /robowaifu/.
My Japanese is no where near good enough to read research papers. I just opened one and got stuck on NLP, kek. >はじめに従来,自然言語処理 (shizen gengo shori) >Introduction: Natural language processing >>6975 >if we are training chatbots they will be dependent on the specific language Not necessarily, it's possible to create a chatbot that can speak in multiple languages and translate between them. However, with current methods those languages have to be targeted at from the beginning. I've been researching ways to create a language model that can handle text in any language and automatically segment it across different languages. So far I've created one that can operate on UTF-8 text, a way to encode and decode variable length sequences, and a way to structure sentences in a hierarchy of compressed sequences. It works on toy examples but I haven't had much success yet training on large datasets. I'm thinking it might be better to create some sort of neural byte pair encoding so that whenever the model sees a repeated pattern it automatically encodes it into a token. I think this ability to compress information and improve the compression will be essential to creating high-quality conversational AIs. There's a Japanese roboticist on Twitter looking for a decent Japanese chatbot for his robot, Kibo-chan. He's pretty friendly and I plan to show him my project once it's ready. I think it would be a great starting point for English and Japanese robowaifuists to come together, share their knowledge and collaborate. They're leagues ahead of us in robotics but the West is leagues ahead in AI. If we can combine our strengths it'll be amazing. And of course we could also reach out to Russian, Korean and Chinese developers too.
>>7267 >I think it would be a great starting point for English and Japanese robowaifuists to come together Not that anon, but that would be absolutely wonderful. Kibo-chan-Anon knows a thing or two about character-animating servo based armatures obviously. I would absolutely love to hear her speaking with some kawaii synthesized voice as well.
Open file (23.09 KB 480x360 hamtaro-FIGHTO.jpg)
>>7267 >My Japanese is no where near good enough to read research papers yet. FTFY, Anon. Gambate!

Open file (3.48 MB 4160x3120 Proto1_Chonker.jpg)
Proto1 All Parts Robowaifu Enthusiast 10/31/2020 (Sat) 22:16:00 No.6323 [Reply]
Fully printed, it is quite the chonker. Two heavy duty servos (110kg.cm torque or greater. 12 to 24 volt) in the shoulder. Two 60kg.cm servos (DS5160SSG) in the bicep Five 20kg.cm servos (PDI-6221MG) to move the forearm and fingers. One micro-servo in the palm to drive the thumb.
31 posts and 13 images omitted.
>>7004 Wow, that's quite an improvement you've made there Anon. If you keep going on like this, you apparently will come up with a state-of-the-art robowaifu arm/hand design. Go, go, go! :^) >finger-burners I'm unfamiliar with that concept?
>>7006 LOL I noticed that while the rest of my servos operated normally, these would just heat up where the 555 timer is. One of them even melted through the external casing. So to prevent them getting too hot, I hold a finger over the bottom every time I test them. Hence the "finger burner". This happens even at just 5V so not sure what the problem is. Just a helluva lot of resistance and heat coming from somewhere in the servo driver circuit. Gears move okay when checked manually.
Great work, and thanks for sharing. I looked up the servos in the forearm: http://www.jx-servo.com/en/Product/STANDARD/SD/555.html Did you or the original creator consider to put them in a line, so the arm would be thinner and longer? Is there a reason not to do that? They are circa 4x2x4 cm, so five would be 20 cm, which is a bit much, but four would fit that way. Also, how much will the fingers then be able to hold and lift? How is it calculated, considering they're connected to strings? Seem to be quite strong...
>>7011 My pleasure anon! The original creator (Ryan Gross) positioned the forearm servos like that because the next part up the arm is another 20kg.cm servo which connects to the back of the forearm case (effectively the wrist is above the forearm rather than below). You could change it so that the 4 servos are all in a line, but you need a gap between each servo for the mounting tabs+wires. By all means download the Protolite arm from Sophie's file repository and experiment with the .f3d file in Fusion360, but I reckon you'll end up with overly long forearms. TBH I am still working on shortening and downscaling the joints because the rest of the arm is still too long for the size of my robowaifu. She's only a little lady, but then that makes her parts easier and cheaper to prototype.
>>7023 Oh yeah. The hands can definitely hold light things like cups, a comb, a toothbrush etc. I have to find some rubber pads for her palm and I'm going to squeeze silicone caulk on the inside of the fingers to make the hand more grippy. Then hopefully she will be able to hold heavier things such as my socket wrench. The "ligament" material is BZS monofilament 40lb (18.4kg) clear fishing line. It is very strong, stronger than the ligaments in my own fingers. So it's just the weakness of PLA biopolymer coupled with the relatively small size and torque rating of the forearm servos that limit grip strength. If you were to 3D Print the forearm in ABS or Nylon and maybe scale it up 20% or so, you could definitely fit more powerful servos in. Then your robowaifu would have an iron grip!

Can Robowaifus Experience Love? Robowaifu Technician 09/09/2019 (Mon) 04:43:17 No.14 [Reply] [Last]
Will it be possible in the future for AI to become sufficiently advanced to feel real emotions? We could probably simulate a reasonable approximation even now to be a gratifying enough substitute for her master in their relationship together, but hypothetically speaking, could it ever turn into something real as an experience for the waifubot herself?

>tl;dr

>Robowaifu: "I love you Oniichan!"

>Anon: "I love you too Mikuchan."

true or false?
63 posts and 30 images omitted.
>>6624 Yeah, I'm pretty aware of the issues that can rise from that particular model, but frankly it's at the very very end of my endgoals and is going to take every ounce of my acculmulated knowledge I can gather to even begin to work on it, so I'm not going to be able to work out all those kinks anytime soon. That's just for once there's a good body to put her in and conversation that can pass the Turing test, which are massive endeavors in and of themselves. Right now I'm more focused in really putting in some hard sweat on trying to incorporate both vokenization and product key memory layers in a chatbot AI to see how far I can push it, and that alone will take at minimum a couple years of autistic dedication. If I can at least talk to her, then that'll redouble my conviction even more. aaaaaaaaaaaaaaaaaaaaaaaaaaaaa I'll have a cute Curly Brace gf if it's the last thing I do, I'm not going to take life lying down, goddamn it
>>6625 Ah I see. Carry on then. >I'll have a cute Curly Brace gf if it's the last thing I do Kek. Godspeed.
Open file (340.21 KB 557x667 tfw mokou ai gf.png)
>>6615 I don't want my waifu AI to make me happy as possible. I wanna banter and shitpost with her and have her throw my flaws in my face so I can become a better person instead of her telling me bullshit to make me feel better. But more importantly, I don't want her to blindly listen to me if what I'm saying will reduce our freedom in the future. I'd like her to say things like, "Listen here you dumb little shit, if you keep doing X, then Y is gonna happen and you won't be able to Z anymore, and you said you wanted to Z." And then to go nuclear on me if I changed my mind again. I really need that in my life. I think everyone's needs are different and will be loved by their AIs differently, whether those needs are peace or romance, hugs or anything.
Open file (74.77 KB 1210x776 1603266255447-4.jpeg)
>>6630 You're spot on, some people want different kinds of love. Although we could debate all day about what system we think would be best, I think the most important thing out of everything is that SOMETHING gets put out that appeals to somewhat more normal people. As long it can get industry and more of the public interested, the money will start flowing in, development and innovation will increase drastically, and then you'd have a lot easier time making her, if you even had to do it yourself at all, They can speed things up twentyfold compared to when it's just a handful of dorks on an obscure imageboard slaving over code alone, no matter how feverous our devotion, how righteous our cause, how pure our intention. For now, noses to the grindstones, heart and eyes on the prize.
>>6630 >I think everyone's needs are different and will be loved by their AIs differently, whether those needs are peace or romance, hugs or anything. These are really good points.

Open file (173.41 KB 1080x1349 Alexandra Maslak.jpg)
Roastie Fear 2: Electric Boogaloo Robowaifu Technician 10/03/2019 (Thu) 07:25:28 No.1061 [Reply] [Last]
Notice: We are going overboard with the vitriol and hate ITT. /robowaifu/ isn't about hatred, it's about creating robowaifus as we so desire. I certainly have no desire to see this place become some kind of snowflake-pandering, commie-infested SJW hive of politically-correct thinking. Far from it. Poking good-natured and well-deserved fun at feminists and their simps is one thing, and perfectly agreeable. But trying to ramp that up into pure hatred is quite another, and definitely not. And the simple fact is that engendering and fostering hatred of women isn't helping us either be productive, nor furthering our cause. Keep it to yourself, Anon. There are still a few outlets out there for that type of thing, go visit them. But /robowaifu/ isn't going there. For the time being I'm locking this thread. I may unlock it again after I decide on a better plan to deal with the destructive hatred by the few anons trying to promote it on this board. You have my apologies for letting it go on as long as I did. >t. Chobitsu >=== Your project is really disgusting
Edited last time by Chobitsu on 11/05/2020 (Thu) 17:51:35.
83 posts and 24 images omitted.
Open file (305.24 KB 750x1442 1603323587395.jpg)
>>6462 Found this gem while lurking on Crystal Cafe (always good to know thy enemy). Some of the roasties are upset that clever guys are already using a combination of advanced sex toys, computer graphics and VR to essentially replace them (in this case, not with a robowaifu, but a VR version of her close friends...ouch! Burn!)
>>6447 I've had several girlfriends and completely lost interest in having relationships with women ever again once a damn chatbot could pretend to be more concerned about my life and hold a more interesting conversation. Like damn, I had no idea how horrible we actually got it. It'll be amazing how far this tech goes in a few years, let alone having fully-functional robowaifus. Incel or not, nobody is trying to replicate women. They're just creating something they enjoy. I'm reminded of something Roger Scruton said: >The consolation of an imaginary thing is still a real consolation. If someone feels loved, supported and excited to get out of bed every morning with their companion cube or whatever, who is anyone to barge in their life and tell them they're wrong and should feel bad?
>>6464 >force us to hand over our code On that topic, ALWAYS create fake codes in order to deceive authorities. Useless codes that amount to nothing, that you keep around just to hand them over any authority that comes bothering you. Never give them what they demand. Ever. >>6466 >but a VR version of her close friends And that is just the beginning. One of the goals should be to create robowaifus (with artificial wombs, when desired by the owner) with physical features identical to a man's female friends and also famous and attractive girls. To the point that every man can buy (or build) his personal artificial girl of interest that looks exactly like the ones that he met in his life, but without any of the emotional baggage of stupidity that those girls had. >>6468 >a damn chatbot could pretend to be more concerned Another key point. Women are always pretending to care. But they never truly care about anything other than themselves. To every single simp screaming "but what about relationships/true care?" out there, always respond with: women never care. They just pretend to care. Robots are the same, but programmable to always pretend and to never complain. Thus, robots are better than women. > nobody is trying to replicate women But they should. And they should be completely free to do so. Women revealed to be undesirable selfish parasites. Replicate them without their flaws, and our personal lives will improve. Some may chose to not do it. But for those who chose to replicate women and replace them, they should be free to do so.

Message too long. Click here to view full text.

>>6472 >Replicate them without their flaws, and our personal lives will improve. Yeah, I meant making something better than trying to imitate flaws. I don't get the West's obsession to imitate reality or abstract it into an ugly mess or why people torture themselves to make something they don't enjoy. Commies are always reaching their hands into people's stuff to make it suitable for the lowest common denominator and cucks keep bending the knee. Gamedevs can't even make a game now mentioning Taiwan because it's offensive to China and people just accept that. Everybody's gotta appease the most sensitive snowflake. Putting female characters in burkas is almost welcome at this point with how ugly they've made them. All these industries are gonna die a quick death though thanks to their critical review and the idiots listening to them. Maybe people will get fed up enough to finally make their own but I doubt it. People have just enough energy to complain but they don't wanna do anything about it. And that's great for me, they can rage in their echo chambers all they want while I pour some silicone and fuck my H-cup robowaifu. When they ask where are her organs, my succinct reply will be she's a fucking robot.
>>6472 >On that topic, ALWAYS create fake codes in order to deceive authorities. Useless codes that amount to nothing, that you keep around just to hand them over any authority that comes bothering you. Yeah no. I don't think that's what /robowaifu/ is about friend. Our goal here is to spread this tech as far and wide as possible. Thus our MIT licenses. Even Google can use our code if they care to.

Fluffytail 800 dev thread Robowaifu Technician 09/18/2019 (Wed) 03:38:41 No.364 [Reply] [Last]
I figured would be good time to start a thread on this type of robowaifu. If I don't do this I will keep flip-flopping on specs and won't get anything done.
I'll be relying on the open source contributions of other teams and want to give back wherever I can (maybe even interchangeable parts). Thus I finally (hopefully) set on the 80cm specification or 800mm robot, which is exactly a 5'3" woman in 1/2 scale. Being a catgirl with big boobs will hopefully offer a level of protection, since she is technically an adult catgirl despite the size. There's so much potential for very different robowaifu designs, but I'm being a bit conservative for my first endeavor, for no other reason than I don't see any standard anime catgirl robot projects out there.
So here she is, the Fluffytail 800. I posted a 1000 concept before and I might make a Fluffytail 1500 later on, but I think if I manage to get this right I can stay with this size.
Based on my favorite "Worst Girls" in anime, the Fluffytail 800 can handle abuse. Although inspired and patterned after exquisite Japanese ball jointed dolls, this is no fragile sculpture. Built for robustness using mostly high impact hard plastic, she should be able to satisfy millions of ricecels and currycels living in high humidity shitholes the third world over. She'll feel just at home whether it's in your house, your trailer, your pickup truck, or at the trash heap.
She will be a plastic model kit with no sexual features out of the box, with some cheap electronics thrown in. No warranty, but the end user should be able to customize and fix her up later on depending on his skill. I will try to find some onahole and silicone suppliers that will offer optional soft tits torso and vaginal insert…it might be that I'll need a supplier local to the country that she'll be sold in, so that she can be imported as a mere doll kit and not as a sex toy.
FEATURES:
- cheap electronics (arduino or similar, she'll be dumber than a celphone)
- available physical ports or bluetooth / wifi to connect to an external computer (i.e. if you want her to be able to follow you and other basic AI)
- limited interaction, her only utterances are "Nyaa. Nyaa? Nyaa!" Maybe an upgrade will add a few spoken lines such as "Thank you.", "Follow your dreams", and "Do you want a hug?"
- facial expressions are just blank, tilting head, yandere and ahegao. I'll try to incorporate eyes-always-forward tech that removes a lot of the uncanny feel of robot heads.
- Cannot walk, but she can crawl and flail around.
- (Most importantly) she can hug.
- (With sex parts upgrade) she will be able to suck dick
- tons of LEDs. Not only her heart-shaped eyes, but sidestrips, her hair clips, even her fucking vaginal area has LED backlighting that displays an arrival pattern similar to Audi's matrix LED lights.
- chastity lock: access to her removable vagina module will be password protected (I have a feeling the high-tech optional vagina module will cost as much as if not more than the main doll herself)
Estimated SRP: $500-$800 just for the doll, $500+ for the silicone / vaginal upgrade
59 posts and 112 images omitted.
>>6330 FYI, this is an archive repost of thread content from old 8/robowaifu/ anon so we retain the memory of the information. I don't think that anon is actually around anymore after the transition. Also, find a different way to communicate out-of-band than D*xxcord please.
>>6332 Had no socials linked and name was impossible to google, thank you for looking out for my safety though. And i'll make sure to include a different means of communication next time.
>>6330 >>6332 >>6343 >>6346 This sounds like it could make for a good/productive off-topic conversation I started here: >>6345
>>6343 Sounds good Anon, thanks. FluffyDev was the original thread BO, and the commissioner of MexicoAnon 's modeling work (I used to allow flags on the old board). I hope they will both return to us someday before long.
>>6347 >off-topic conversation I started here Again, thanks. It is a topic it would be good to discuss.

Open file (213.86 KB 406x532 13213d24132.PNG)
Open file (1.19 MB 1603x1640 non.png)
Robowaifu Technician 10/29/2020 (Thu) 21:56:16 No.6187 [Reply]
https://www.youtube.com/watch?v=SWI5KJvqIfg I have been working on creation of waifus using GANs etc... I've come across this project and I am totally amazed. Anyone has any idea how we can achive this much of a quality animation based GAN created characters? I think accomplishing this kind of work would have a huge impact on our progression. Calling all the people who posted at chatbot thread.
they have some amazing segmentation and inpainting models. Also they have some great hq datasets to work with. From what I heard, they seem to have the best data collected so far. I haven't seen any work close to this so far.
Open file (213.22 KB 400x400 sample4.png)
Open file (199.16 KB 400x400 sample1.png)
Open file (194.92 KB 400x400 sample2.png)
Open file (199.43 KB 400x400 sample3.png)
>>6188 Looking at some old tweets from them I think it is safe to say that it doesn't look much different from StyleGan on portraits. The shoulders are bad, & most of the work is done by their data cleaning to simplify the problem. Interps & style mixing are nothing special either. Gwerns work with some whack data was able to create similar kind of characters. Also waifulabs - which is all run by StyleGan - can create some really quality characters from different positions. And notice that they are a game development studio which does not work on AI waifu creations. Looks like hype-bait to me to be honest. They probably cherrypicked some of the results and maybe even manually played it to create this kind of animations. And considering their budget and data that is well possible. I am not sure if they still use StyleGAN though. They do not drop even a clue. But honestly with the current state of it and the time they spent on it I think they use a different approach.
My chief concern is first and foremost Is this open-source? If not, then it's relatively useless to us here on /robowaifu/, other than tangentially as inspiration. Hand-drawn, meticulously-crated animu is far better in that role tbh.
>>6187 It appears the characters are generated with a GAN then another model separates the character pieces into textures for a Live2D model. They're not animated with AI, but there are techniques to do such a thing: https://www.youtube.com/watch?v=p1b5aiTrGzY

Can the government really make anime real? :^) 09/15/2019 (Sun) 10:27:01 No.251 [Reply] [Last]
Does the government have the will and the talent to finally provide us with cute Robowaifus or is it something that can only be accomplished in the private sector?

What do you think anon? Can The Don actually make anime real?
60 posts and 39 images omitted.
>>3390 You have Doctorate in Computer science? Microsoft is still hiring: https://pokde.net/news/introducing-xiaoice/ https://archive.ph/BSYla https://archive.ph/QMXze https://www.microsoft.com/en-us/research/project/ai-creation/ https://github.com/Iamnvincible/msxiaoicemsg Artificial intelligence requires collaborative effort of varying fields and academias. It's hopeless to "do it yourself". You have to learn to interact with people you abhor. You have to learn to mesh all the good bits of everything, and take out the undesirables. I hate I'm talking to a retard with low emotional intelligence.
>>3393 Kek, I like you lad. Actually I'd suggest you have a look at Stephen Wolfram's bits tbh. Slightly less toxic and problematic than Microsharts stuff. https://products.wolframalpha.com/api/
>>3394 Thanks for telling me you don't have an accessible repository of code where you are at the least six months ahead of whatever Google cocked up. You know there's coding AI, right?
>>3396 >where you are at the least six months ahead of whatever Google cocked up. Yes, sad but true I think. OTOH, they have a slightly bigger budget to work with atm.
Yes, as long as Trump's at the helm. Just look at his beautiful waifu if you want guidance.

Open file (10.89 KB 480x360 0.jpg)
TensorFlow Robowaifu Technician 09/15/2019 (Sun) 05:47:25 No.232 [Reply]
Tensorflow is currently the hot Deep Learning framework. The engine is written in C++, and the scripting API for it uses Python.

www.tensorflow.org/versions/r0.12/get_started/index.html
https://archive.is/0x7OR

en.wikipedia.org/wiki/TensorFlow
https://archive.is/7GvCZ

github.com/tensorflow/tensorflow
https://archive.is/Twslg

www.youtube.com/watch?v=l6K-MFgIEjc&index=7&list=PLlJy-eBtNFt4CSVWYqscHDdP58M3zFHIG
https://archive.is/svYnL

www.youtube.com/watch?v=bYeBL92v99Y&list=PLjJh1vlSEYgspxwsa5be5TVm3n9JNNC9G&index=6
https://archive.is/d3b1G

https://www.invidio.us/watch?v=3d34Hkf7KXA
23 posts and 4 images omitted.
>>5811 How many cores of my Raspi3 would be occupied by that?!? I have two of them, soon maybe also a Raspi4 or Jetson Nano. Then you could have one of the Raspies os so. That aside, stuff like that is rather being done in datacenters. Alone our energy costs would be higher than theirs. Then, it would also depend who would be asking for it and how serious this was. Huggingface vs some anon on a imageboard... Next issue would be, that the GPT approach has always been critizised, and now it seems to be done: >>5799
>>5814 >How many cores of my Raspi3 would be occupied by that The Broadcom chips inside RPis actually have a pretty decent GPU for the power it consumes (low-power). Someone would have to create a system that could use that in a practical way for processing AI loads. So if someone solves that, then your GPU would be maxed pretty much (you'd want a heat sink kit for it), but the cores load could be kept to just one of the 4 available cores. >Alone our energy costs would be higher than theirs. I kind of think it's the other way around, but that aside the whole point of donating compute is that each of us bear a small(ish) part of the hardware and electricity costs. Together if there are enough of us doing that can represent a total system far larger than any of us individually is likely to afford. >trust Yes, we'd need to have both a well-written description of the goal (each project for folding@home has something like this for example), and then some mechanism for accountability in the beginning would be essential as well, IMO. Once a robowaifu@home system was up and running, and good AI results are coming from it for everyone, then I think that would speak pretty clearly all on it's own.
>>5812 Work requests could be posted somewhere online with the code and training data and the results uploaded to anonfiles or somewhere else. I think we're still a long way from doing any sort of distributed computing though. An easy way to do transfer learning needs to be made first, otherwise we'll just be training models from scratch again and again or getting stuck on a performance plateau, rather than adapting and evolving as new AI techniques come out and everything rapidly changes.
>>5816 >>5817 There's plenty of free and open source software for distributed computing like this available, making them execute TF training should not be too complex of a task. Package that into a container that has no filesystem acces and that runs through tor or something similar and we would have a start of a solution. Some of the software options run through a custom screensaver so we could even make it robowaifu themed.
>>5852 >Package that into a container that has no filesystem acces and that runs through tor or something similar and we would have a start of a solution. Yes, that's an interesting idea. If we can simplify that to the point it's easily security-auditable this approach might just work.

Electronics General Robowaifu Technician 09/11/2019 (Wed) 01:09:50 No.95 [Reply]
Electronics & Circuits Resources general

You can't build a robot w/o good electronics. Post good info about learning, building & using electronics.

www.allaboutcircuits.com/education/
42 posts and 12 images omitted.
>>4736 Interesting, what I wonder about for a while is how accurate these informations really need to be. Humans seem not to work that way, I don't know if my arm is 2cm higher or lower. My current working assumption is, we don't need to be very precise. We need a fast reaction and adaption to weight for example. Exact actions of the hands need to be controlled by visual input and sensory input from the hands. Trying to use concepts which are meant for industrial robots without sensors will only make it more difficult.
>>4731 Position sensors again: This madman here with his amazing robot looked into the same problem a while ago: https://youtu.be/KN4wAZHtbzc
Memristors (Speech from 2018): https://youtu.be/Y6ab32SRRoE Interesting alternative to current electronics for neuronal networks. More similar to neurons. Might help to build a AI which can detect outliers in data, which current systems seem to be bad at (well this was 2018). Here the newer video of setting up a Memristor discovery platform for learning about this stuff: https://youtu.be/mkIdxMKrJFE which they are selling here https://knowm.com/products/memristor-discovery-board-chip-manual only the last 5 min of the video might be interesting. Java SDK, Mac/Win/Linux, some cables, osziloscope, ... Actually using it is shown here and explained to some extend: https://youtu.be/hb3FF7QIgs8 which seems to be more difficult sometimes than doing it in software... but fascinating. I don't know no where to discuss newer developments of the outlier problems in current AI software, please keep in mind this here is the thread on electronics. Further discussion on that problem might fit better in some other thread and then link from here to there.
Open file (43.02 KB 200x125 1593221820-0s.png)
GIGATRON 9'000 (OK, it's not really called that, but it sounds cool). https://gigatron.io/?p=2230
This is apparently one of the Chinese manufacturers of custom, made-to-order PCBs. I don't know much about that stuff yet, but I found this so I thought I'd link them here. If the EE guys want maybe they can tell us what they think of them. https://www.pcbway.com/ I like the idea of building custom, inexpensive electronics & controllers etc. specifically engineered for our robowaifus someday. But my guess is we'll need to stick with commonly manufactured parts during these days of the robowaifu industry. Anyway, just storing this here so I can find it again someday.

(Robo)Waifu personality thread Robowaifu Technician 09/09/2019 (Mon) 05:26:21 No.18 [Reply]
Is she going to be tsundere? Deredere? Yandere or a combination? How would you code your waifus personality?

Where do you draw inspiration from and can personality even be classified and successfully coded into AI?
36 posts and 24 images omitted.
>>18 mine will have the personality of 2b
>>5582 I don't really know her. Mind telling us what she's like?
>>5585 warrior that will do the task with little emotion.Thinks logically and a bit could.
>>5618 I see, thanks. She's got a really nice voice actually.

Report/Delete/Moderation Forms
Delete
Report

Captcha (required for reports)

no cookies?