I fucking hate toxic positivity. Every fucking corporation pushes the notion that "lifE iS aWeSomE, wE cArE abOuT pEoPle" and other such bullshit, and when you point it out, they call you a bad, toxic person.

No, you don't care about your community, let alone the whole world. You're just trying to make people believe that spyware, wage slavery and being fired by a neural network is the norm. You're making money off of those who don't have a choice.

If you account all people, not just American white rich 1%, it turns out that for the vast majority of people life is either an uphill battle or straight up nightmare. People are working in shifts and have no time or emotional resource to spend on themselves. Most of the people can't afford a house or a flat. Even those who can still suffer from mental illnesses, to the point where there are more mentally challenged people than mentally healthy ones. The word "neurotypical" meaning "mentally healthy" is wrong.

You want nothing but to sell your stuff and earn more money off of Chinese and Indian factory workers who work 16-hour shifts. Maybe your life is great, but aggressively pushing this notion is a big, wet spit in the face of humanity.

Fuck you. Fuck your space rockets. Fuck your twitter accounts. Fuck your institutionalized exploitation of the weak. Fuck your products. Fuck your "open source". Fuck your "GDPR compliance". Fuck your offshores, your hedge funds and your tax evasion. Fuck your bailouts. Fuck your ships spilling tons of crude oil, fuck your factories, fuck your slave labor, fuck your anti-suicide nets in Chinese dormitories.

One day, because of you, our planet will become unlivable. You will hop into your fancy space rocket to go to that top-1% elite Mars colony. Nice job.

But I will pray for a solar flare to hit you and turn you and your fucking rocket into radioactive ash.

I hate having a 12 hour shift at subway after studying neural nets at home...

Round up kids.
I have a story to tell. The story of a war I've lost. Many battles were fought and many hours were wasted.
This is the story of wasp in a computer lab.

Today, the weather was good. So your old pal, Nomi, decided to open the windows. And as usual, that's where it all started.
So Nomi sat down and worked for a few hours. Tweaking two different neural nets, adding to its dimensions and concatenating the living shit out of the data they were supposed to process. After, she tried testing and testing and testing. It was early afternoon at this point and she was hungry. She went to close the windows and go for lunch.... When she realized, that she's not alone in the room. A big ass wasp was sitting on one of the curtains.

Now, Nomi doesn't have a good relationship with bugs and flying shit. Wait, no, she doesn't have a good relationship with moving things in general. So she panicked. She begged the wasp to leave. The wasp sat on the curtain and smirked at her. So after a while, she left the windows wide open, turned off the lights, put her hoodie on and went for lunch.

(btw, at this point my hoodie smells of sweat, fried onion, steak, cigarette and shisha. Don't ask. It was a long two weeks)

When she came back, the wasp was nowhere to be seen. So she assumed that the wasp got tired and left. But oh, how wrong she was.

After few hours, she heard something. She assumed it was just a fly. Actually, she hoped it was a fly and not the return of the wasp. But all her hopes were in vein.

She heard a buzz. And all of a sudden, an angry wasp flew in her direction. She dodged the attack and got under the table. But the wasp was not letting this go. Nomi jumped out of the room and left the door open. The wasp hid itself. She waited and waited but no sign of wasp. So she ran back in the room, and opened the window and ran back outside. She waited. The wasp occasionally would fly from one hideout to another. The wasp was making herself comfortable. At one point Nomi got angry and threw a shoe at the wasp, but the wasp caught the shoe and threw it back at her while maniacally laughing at her.

So she gave in. This was enough for the day. She ran back in, closed the window, turned off the computer, took her bag, turned off the light, and closed the door. All in less than 15 seconds. She came outside panicked and distressed, and now she's on her way home hoping that by tomorrow the wasp is gonna be dead.

The wasp and the robots are sitting alone in the lab tonight. I hope when the robots uprising happens, the robots can forgive me for abandoning them powerlessly with a wasp. 😟

Fucking facebook researcher that make underfitted neural nets and fuck Mark that it's a marketing genius, the only idiot that can make news from a failure. The CEO of Tesla knows it and said Mark is not an AI expert. Bug not feature, it's only a poorly trained and poorly designed neural network having a bad representation of concepts, not a new language and not the fucking apocalypse. Google faced and solved the same issue when start ed using neural nets for zero-shot translations without using english as a translation bridge.

So, as I'm currently cut off from the world of tech, my anxiety in regards to research has settled and I actually enjoy doing it again on my terms. It just makes me jealous to look at all these people developing cool stuff and wanting to get in on the project and maybe improve a part or two, particularly the robot kind. I want to slap some neural nets on majority of the robotic shit I see, or optimize them, or do something to make them more robust... But I don't have a research position right now where I can spend time and money doing that. So I just sit in front of my laptop and sulk.

... And literally this is why we can't have nice things. Cuz I'm not hired to make nice things. Literally.

1.
Accuracy 0.90 achieved so easily, makes me wonder if I've done something wrong. Lol.

2.
My neural net models are the only things in my life doing well. I think I chose the right career. Lol.

3.
Rerunning experiments is not fun. But getting better results is really... Ego stroking.

i7, 8GB RAM with 2GB nvidia graphics couldn't handle the training of neural net on a google text corpus! -.-"
I'm just watching it train, nothing to do! oh but wait,I can rant! 😃😃

Took a class on neural nets once upon a time and all the prerequisites had been taught in C/C++ but the professor insisted on teaching in Matlab because they didn't know C/C++.

The first fruits of almost five years of labor:
7.8% of semiprimes give the magnitude of their lowest prime factor via the following equation:
((p/(((((p/(10**(Mag(p)-1))).sqrt())-x) + x)*w))/10)

I've also learned, given exponents of some variables, to relate other variables to them on a curve to better sense make of the larger algebraic structure. This has mostly been stumbling in the dark but after a while it has become easier to translate these into methods that allow plugging in one known variable to derive an unknown in a series of products.

For example I have a series of variables d4a, d4u, d4z, d4omega, etc, and these are translateable now, through insights that become various methods, into other types of (non-d4) series. What these variables actually represent is less relevant, only that it is possible to translate between them.

I've been doing some initial learning about neural nets (implementation, rather than theoretics as I normally read about). I'm thinking what I might do is build a GPT style sequence generator, and train it on the 'unknowns' from semiprime products with known factors.

The whole point of the project is that a bunch of internal variables can easily be derived, (d4a, c/d4, u*v) from a product, its root, and its mantissa, that relate to *unknown* variables--unknown variables such as u, v, c, and d4, that if known directly give a constant time answer to the factors of the original product.

I think theres sufficient data at this point to train such a machine, I just don't think I'm up to it yet because I'm lacking in the calculus department.

2000+ variables that are derivable from a product, without knowing its factors, which are themselves products of unknown variables derived from the internal algebraic relations of a product--this ought to be enough of an attack surface to do something with.

I'm willing to collaborate with someone familiar with recurrent neural nets and get them up to speed through telegram/element/discord if they're willing to do the setup and training for a neural net of this sort, one that can tease out hidden relationships and map known variables to the unknown set for a given product.

by simply making the bias random on the second input for a two bit binary input during activation calculation, it's possible to train a neural net to calculate the XOR function in one layer.

I know for a fact. I just did it.

Continue making my app, or start on neural nets?

Neural nets hiding some images and recreating others

I am at a work seminar and the presenter is talking bullshit about artificial neural nets.

Unfortunately I can't punch him through the webcam. This is frustrating. Why do morons who know nothing about neural nets always insist on talking about them?

Some friends of mine were working on doing neural network image processing and wanted to build a social network for it. I got to play with graph databases, mobile app development, and neural nets. Unfortunately, project never took off, but it was fun nevertheless.

Could neural nets be used to solve a complex problem with a lot of predefined specific weights and biases related to real activities that could be numerically represented ?

Like if the various layers represented things who’s outcome would be cascading and a direct outcome of factors inputted to them resulting in limited number of outputs ? Like say maybe wind flow at locations resulting in a wind current at an expected time in another location and where a system would have to change to in order to result in the final expected i output ? Maybe a bad example as that could be affected by a lot of things .

But basically the gradual massaging of values that would relate to causal effects where a specific output was designed being intermediaries between the desired input and output ?

What ai model would I use to propagate a series of survival factors and decision making scenarios that if the optimal order of activities are pursued would lead survival and even prosperity and the worst set of possibilities would lead to death where the environment and sensations being experienced would always lead to specific pitfalls but wherein some of these pathways would lead to later reward and where the obstacles like predators could be overcome by simple combinations of objects which would be a crude mimicry of the invention ?

Neural nets don’t see to fit this given my understanding but there is a training aspect I’m looking for where the creature being simulated dies, develops fear responses, feels pain, avoids pain, remembers things, develops behaviors related to characteristics in creatures, has unborn motivations that weight decisions, and learns to prioritize.

I had created a massive dataset of objects including memories and aspects of semantic memory and episodic memory colored by emotion inspired by past conflict and reward with the idea that a running average would affect behavior and decide on various behaviors all the way down to perceptual differences

Any thoughts again ? Or will wolf try to steal these too ?

How do they parse and arrange the input and then generate the output with neural nets for talking bots?

Any suggestions for a good starting point for learning to do more with neural nets? Not interested in image recog so much, but would like to see the cutting edge of textual pattern recognition... I dunno, I don't even want my expectations to color this... whats do you guys find most interesting and enjoy playing with? Python is preferred but I'm grateful for any tips/links/ideas/rants you might share!