absolute peak🙌 we need part 3 ASAP!!!

This is the foundation model I was searching for to mimic cosmo-biology dynamics (energy-function astrology) thanks!

Incredible content honestly

best video on creating a simple workflow that also works great for students and researchers! i find that most workflows require more time in decisions and thinking about structure more than just doing something.. this is bare bones and that‘s great! thank you <3 i also study neuroscience and it was hard in how to separate lectures/courses from self studying things / independent research. this is great!!

Спасибо за видео. Простите, у Вас канал на русском? Ваша имя и фамилия русскые, поэтому я думал, что вы также говорите по-русский

Bravo! An excellent explanation. Extremely useful. Thank you.

think turing engine. problem: how to transform a brain into a turing engine.

Nice explanation.

Such a gem of topic. Unselfish land so enthusiastically produced. Thank you. Foe putting this together for the interested. ❤

My brother who is graduating high school this year recently said to me that he might do neuro science in uni because he feels like the brain would be easy to study. It was at that moment that I realized my brother does not have a brain

Isn't the probability of all the gates being open being n^4 implicitly assuming that all the gates act independently from each other?

That tattoo is the coolest thing man

there is no equation of neuroscience, eff you and your stupid accent

its a dreaming machine 🤯

You are just amazing!

Your stock resources, like from where you download assets for your animation? Thank you 👍

You're an amazing teacher. This was very insightful, well-structured, and wonderfully presented. Reminds me of 3B1B. Keep it up!

this is how to turn capacistance recharge and discharge into equeations and then someone claim it has to do with intelligence and neurons. you can try with analog electronic components and network a bunch of capacitors with transistors or vaccuum tubes and see if you can make these interaction in some intelligent way that can mimic just the action potential of the cell to emulate analog cell but it would never work cause there is something missing. the cell is not just creating a voltage. there is chemicals molecules created that is send from axons to dentries by the action potential that regulate the metabolism of the cell that then produce chemical molecules that then trigger another transmission. that is what controls the firing in the first place. there is nothing quantum about this.

but real neurons are just nerve cells that can communicate in arangements that arent even in a network arcitecture. real neurons can be arranged as a long string of cells but still communicate more intelligent that the artificial network. its because its not the network that make it intelligent. its the way the cells communciate itself that makes the real neurons more intelligent. perceptrons are just adders. they add numbers togheter. its not thinking. biological neurons uses symbolic language. they dont send the same message from cell to cell. they send different message. its not the electric pulses that make the biological neurons intelligent. its the chemical molecules that is transferred trought the ion channels that contain that communicaiton language. the voltage is just a mean to transport the chemical messages to the receptors that contain the thinking stuff. just as a fractal the timing of the firing pattern can contain a show image of the communication of the chemical messages cause its the chemcial messages that regulate the firing pattern. its the chemcial messages that program the timing of the electrical pulses beteen the nerve cells. a artificial equal to chemcial messages and action potentials is a set of numbers transfer from on piece of code to the nexrt trough separate channels that allow branching. the reciver code take the first code segment output and use it as input to perform a task then it generate another sequence of numbers to send to the next code segment. each code segments then acts like biological cells. think of the original biological cell as contain receptors that act react to chemicals in a fluid and produce a single cellular behavior to these chemicals in the fluid. when it uses another cell to communicate it emulate the same kind of chemical molecules that it react to in its external environment as a single cell. the next cell in line has to produce the same kind of signals that the first cell received from the fluid environment it lived in that it send directly to another cell but the signals are not the same signals from cell to cell cause each cell has to learn how to operate in a similar way to that of single cells in a fluid environment but in this case they just read chemcial signals and then generate new ones that define behavior of the next cell in line. its the molecules in the chemcial messages that are the programming language the cells communciate that drives the timing of the firing. there is a internal program that in single cells act on external chemcial signals from the environment that regulate the single cell behavior that in nerve cell do the same as external input but with chemical signals created from another cell instead of the fluid environment single cells lives in. this is the only logical probable explnation to biological cell communication. both aucaryote and bacterias behaviors is the basis of explanation to more evovled cells like nerve cells. the intelligence is not in the network. its not in the firing. its in the way the cell learns to produce chemcials signals that act as instructions the next cell in line has to use to act on. think of a single cell as little creature that need to survive in a fluid. some molecules are frendly. some are dangerous. the cell may want to move toward certain moleculas and avoid other molecules. these molecules it get in their receptors defines their behavior and how they react. there is a internal program in the cell that uses the detection of chemcial molecules in the receptors of the cell to regulate the behavior of the single cell. the same regulatory mechanism is invoved in nerve cells that define the behavior of the nerve cells when they receive signal molecules. the difference of single cells and nerve cells is that nerve cells can create and emulate the same kind of input data that the single cell receive from its environment to act as trigger mechanism that the internal program in the nerve cell uses to deploy certain actions. this is a far better explanation that any one else would give you about biological cell. its the metabolism of the nerve cell that produce action reaction when chemcial messages are being read. its the internal programming of the metabolism of the cell that reads the chemical signal molecules and produce internal processes from that. the cell would act differently on different arangements quantites of chemcial signals. put it simple. just like a assembler write source code into machine code the cell metabolism reads chemcial signals into actions in a similar way. its the chemical signals that regulate the metabolism that then encodes new chemical signals that then regulate the metabolism of another cell. one cell learn how to relulate the metabolism of another cell. the other cell then has to learn how to regulate the metabolism of the next cell in line. if your intelligent you get the picture. this is very different from percentrons and their number adding networking as biological cells are little creatures that has learned to function in a collective as a single creature of many cells.

God bless the calsequestrin that got inhibited to release calcium ions which made conformational changes to troponin that made the lumbricals and interosse work and write the algorithm that recommended this video to me with the help of prefrontal cortex and the motor cortex area in the parietal lobe that recommended this video to me , special thanks to all the ions and energy sources .

its thru time!!! not over time brat... time is present at every scale..

00:01 Back propagation unites all machine learning systems 02:44 Backpropagation is essential for training a network 08:05 Differentiability enables efficient knob adjustments 10:36 Derivative of a function is the instantaneous rate of change 15:54 Iteratively adjusting knobs in the direction of minimum 18:32 Gradient descent is an iterative procedure for minimizing a function. 23:16 Derivative of a combination of two functions with one as input to another 26:44 Using chain rule to find derivative of loss function with respect to input knobs for iterative gradient descent 31:52 Gradient propagation in a computational graph 34:17 Derivative propagation in machine learning Crafted by Merlin AI.

Super great explanation !

This was amazing, thank you so much! I would LOVE to see a video on VAE's / ELBO or the various ways we can use generative models by trying to learn the joint distribution.

Great video, thank you very much , note to self(nts) watched …… 31:25

Thank you❤such an extraordinary presentation with relations and simplicity

This is unbelievable, I loved it, now I can see how the KL divergence can be used to correct our approximate models, and how we aim at minimizing this divergence to get closer to the hidden distribution, beautiful!

Thank you Artem, this is amazing!

beautiful, both the information and the guy

Excellent video!!!

Thanks!

Brilliant, thank you! NY finest!

Around 15 mins, we get that the prob of a 4 gated channel being open is n^4. However does this not require an assumption of independence between gates and would it really be appropriate to assume independence here?

We'll information good show and 😅😅

AI music is NOT there yet tho. Like its fine for like background music in corporate ads but i have yet to see a piece of ai music that i would listen to in my spare time.

hahahaha he assumes people remember dot products from high school. i know civil engineering undergrads that couldn't tell you about it the second they're out of statics. MURRICA

Flat Earther! Kidding hehe

Dear Artem Kirsanov, I hope this message finds you well. I recently watched your KZbin video, "Wavelets: A Mathematical Microscope," and I must say, I was truly amazed by the clarity and elegance of the animations and graphics you used to explain the concepts. The presentation style really helped me understand the subject, and I can tell a lot of care went into creating it. I'm very interested in learning more about the tools you used to make these visuals. Could you please share which software or techniques you used to create such engaging animations and graphs? I'm currently working on similar material, and your approach would be incredibly helpful for my project. Thank you very much for your time, and I look forward to your response. Best regards, Mohamed

There isn’t such course called vector calculus in brilliant that is accessible to me. Is it my device’s problem? Please help me……

Great explanation!!!

One of the most precious channel on KZbin

subscribe just by watching you explain math

What determines the equilibrium potential? Why would the it not be zero?

I genuinely think it's a form of proof of evolution, that IS what is happening, on an electronic scale, but still.

I have never had a video blow my mind as much as this one and I have no background in physics or neuroscience. Congratulations !

🤓 It's the gay science! 😆 10:54 🤣 I = gay (Erect - Vagina) 😂 10:56 😄 I = gay bar person (Erect - Vagina) 😁

Excellent explanation!

Have you ever been so far as go want to do look more like? I have been farther and wanted more than ever has even decided to use!

Just WOW. Thank you so much for the great job. This is indeed a work of art. This is indeed a work of art; you convey so much information in such a short time and in the best manner. Please do not stop making such videos.

What program is used to create such a beautiful presentation?