How AIs, like DeepSeek, are so effective

[@DanielBoctor]

To try everything Brilliant has to offer -free- for a full 30 days, visit 👉 brilliant.org/DanielBoctor/. You'll also get 20% off an annual premium subscription! THANKS FOR WATCHING ❤ JOIN THE DISCORD! 👉 discord.gg/WYqqp7DXbm 👇 Let me know what type of content you would like to see next! 👇 Thank you for all of the support, I love all of you

[@MaakaSakuranbo]

"you clicked on this video because you wanted to know more about neural networks" Actually 3x+1 just reminded me of the collatz conjecture

[@lindhe]

Yeah, I was completely click baited by that! 😅

[@machine-boy]

same I got baited by it too

[@w花b]

Right?? I was expecting some link between the conjecture and NNs :(

[@senyai]

I feel Rickrolled

[@LambdaJack]

@15:35 It was intentional. I don't see the point. Exactly as _advertised_. I was mistaken to expect anything less from the _Brilliant_ title!

[@PascalProducers]

This was the first video that explained neural networks in a way that actually made sense to me- showing me the math concepts behind it, while not going to the actual level of the calculations and becoming overwhelming, which made it really enjoyable and easy to follow, as well as excited to learn more about the reLU/log functions and how they truly work. Thank you!!

[@DanielBoctor]

thank you for the support, I'm glad it was helpful 😊

[@imyourmanzi]

you explained succinctly in 20 minutes what my college AI professor couldn't get across in 15 weeks

[@ASCENDEDPulsar]

I remember trying to wrote basic neural networks in 2019 and got lost in terminology of activation functions. I want to say thank you for giving clear explanation on sophisticated fields, be it reverse engineering or neural networks

[@bradleystone6498]

You really did a great job explaining this. It would really be great if you extended this to show how probabilistic functions are included in this process and how LLMs generate the next token based on the previous chains of tokens.

[@EnginoIsRunning]

Your videos are astonishingly clear in every way. Even with just a basic knowledge of coding, it's possible to understand everything. Good job!

[@Shaunmcdonogh-shaunsurfing]

Very well presented and thank you

[@garbage_blob]

Very informative video. Not my field but you've explained it really well to a layman

[@ataarono]

9:40 negative house prices are perfectly reasonable. If you for example input a given prison cell, it should output the Bail price.

[@mohammeddigital]

Damn!

[@tdv8686]

😂😂

[@Bola382]

clickbaited by the Collatz conjecture

[@jacobjayme6280]

ANOTHER BANGERRR BY DANIEL BOCTOR!!!!

[@airman122469]

Yes. Neural nets are just combinations of linear functions. It’s like a learned Tayler series expansion for nearly any learnable pattern.

[@MartinBlaha]

Thank you very much ❤ I like the way how you're explaining quite complex topics. I'd love to see more on AI related topics for coders. We need to learn how to elevate AI and this starts with understanding of the relevant concepts like neural networks. Well done 👍

[@texastoland]

I already understood NNs but liked the revelation that using it is just like training without back prop 👏🏼

[@royalsrivastava2079]

good work . i really liked the way you explained this topics . its in my syllabus and this was the best explanation. now i can solve numerical easily

[@RealRatchet]

Calling them linear is only technically true because it can't really do proper approximation of a non-linear problem without a non-linear activation functions, what it does is train the slopes of the the linear activation function. But I imagine this will be adressed by the time I finish watching the video.

[@shantanusapru]

Nicely explained! Looking forward to those other 2 in-depth videos...!

[@MrVohveli]

"So it's all functions?" - Always has been.

[@oliver1784]

are we just 4 dimensional neural networks? whatever that means... also, great video! ❤

[@ginebro1930]

Non linear, and can imitate any non linear function.

[@genesisreaper2113]

Weights and biases, neurotransmitters and minimum activation levels, tomato tomato potato potato. All the same thing the further down you go. With advanced enough technology you could turn anyone's brain into an algorithm. Assuming quantum physics can be put into a equation, of course.

[@xanaxity]

it's y=mx+c. 🤔

[@nickthetalldude]

Thumbnail infringes on CGP Gray trademark

[@artemis1825]

3b1b OST?

[@DeGandalf]

Just a very small correction: For the definition of deep neural networks it actually also needs to extract the features on its own. So just stuffing the housing features into an MLP doesn't really make it deep learning.

[@entropycat]

I hate your click bait title!

[@Darkev77]

It’s not though

[@Not_Him_FrFr]

​@@Darkev77they probably just meant the way it attracts people to click, making them expect more, not that the video itself is clickbait

[@LambdaJack]

Too much to expect from the author, an LLM itself, and it's followers... RegExps I guess.

[@TrusePkay]

Video is good but stop using other people's thumbnails and theme music

[@ImperialRoads]

This man is the goat fill me up with more content king 🔥🔥🗣️

[@jtw-r]

I am neural network, I want to learn about human

[@jtw-r]

I think i’m lost

[@zxcaaq]

clickbait trash,