Скачать или смотреть unnamed draft - physics inside neural network - graphical representation

so this is newtonian mechanics, specifically the equation of gravitation combined with a neural grid, which applies the force of attraction between the masses of nodes in the shared space and time.

exact equation is given by: F = G * (m1 * m2) / r^2

above F is the force of attraction between an individual node while m1 and m2 are the two masses between any 2 different objects. G is the gravitational constant which is universal and describes the strength of the gravitational force.

regarding artificial neural networks they are at its core straight forward mathematical equations. while conversely a powerful neural network is an even more complex equation. like in the brain, we connect multiple neurons together and form a network that can be trained to perform tasks.

simplified equation here: f(x) = z(x) = w * x + b

in this example above f is an individual node while z equals result or preferred output value. w is weight defined by training data and x can be any input or output value. b is the bias modified in order to get results.

in this recording, however, we ended up excluding many more complex dynamics, for example acceleration, angular motion and especially collision ^^ since that would create a totally different visual dynamic =) however, this serves as a basic example of how to simulate particle systems both with / without gravity and / or inside a neural environment.

to simulate more realistic physical systems, there are many numerical techniques that can be used such as verlet integration, runge-kutta method or finite-difference method for solving equations of motion. additionally, many physics engines, such as bullet and havok provide ready-made solutions for simulation physics, which includes various features such as collision detection, rigid body dynamics and constraint solving.

化学 - 技术 - программирование - प्रणाली - 코더

➡️ twitch:   / bitrateseemsfine  
➡️ instagram:   / binarystate  
➡️ network:   / binarystate  
➡️ chat:   / discord  

➡️ web: https://web.binarystate.repl.co
➡️ shader: https://tinyurl.com/y3fwkfwb
➡️ youtube: https://shorturl.at/lqy25
➡️ openai: https://chat.openai.com
➡️ qb64 site: https://qb64.com