The Essential Main Ideas of Neural Networks
Вставка
- Опубліковано 17 чер 2024
- Neural Networks are one of the most popular Machine Learning algorithms, but they are also one of the most poorly understood. Everyone says Neural Networks are "black boxes", but that's not true at all. In this video I break each piece down and show how it works, step-by-step, using simple mathematics that is still true to the algorithm. By the end of this video you will have a deep understanding of what Neural Networks do.
English
This video has been dubbed using an artificial voice via aloud.area120.google.com to increase accessibility. You can change the audio track language in the Settings menu.
Spanish
Este video ha sido doblado al español con voz artificial con aloud.area120.google.com para aumentar la accesibilidad. Puede cambiar el idioma de la pista de audio en el menú Configuración.
Portuguese
Este vídeo foi dublado para o português usando uma voz artificial via aloud.area120.google.com para melhorar sua acessibilidade. Você pode alterar o idioma do áudio no menu Configurações.
For a complete index of all the StatQuest videos, check out:
statquest.org/video-index/
If you'd like to support StatQuest, please consider...
Buying my book, The StatQuest Illustrated Guide to Machine Learning:
PDF - statquest.gumroad.com/l/wvtmc
Paperback - www.amazon.com/dp/B09ZCKR4H6
Kindle eBook - www.amazon.com/dp/B09ZG79HXC
Patreon: / statquest
...or...
UA-cam Membership: / @statquest
...a cool StatQuest t-shirt or sweatshirt:
shop.spreadshirt.com/statques...
...buying one or two of my songs (or go large and get a whole album!)
joshuastarmer.bandcamp.com/
...or just donating to StatQuest!
www.paypal.me/statquest
Lastly, if you want to keep up with me as I research and create new StatQuests, follow me on twitter:
/ joshuastarmer
0:00 Awesome song and introduction
2:01 A simple dataset and problem
3:37 Description of Neural Networks
7:54 Creating a squiggle from curved lines
15:25 Using the Neural Network to make a prediction
16:38 Some more Neural Network terminology
#StatQuest #NeuralNetworks #DubbedWithAloud
The full Neural Networks playlist, from the basics to deep learning, is here: ua-cam.com/video/CqOfi41LfDw/v-deo.html
Support StatQuest by buying my book The StatQuest Illustrated Guide to Machine Learning or a Study Guide or Merch!!! statquest.org/statquest-store/
NOTE: A lot of people ask where the values on each connection come from that we use to multiply and add to the inputs. This question is answered at 3:48 - in short, they are found using a method called Backpropagation. If you want to learn more about backpropagation, check out this video: ua-cam.com/video/IN2XmBhILt4/v-deo.html
Bamm
Calling it squiggle fitting mechine is genius
I can't believe I've only just found out you published a book! I just bought it, can't wait for it to arrive! ❤
@@ThisGuy0903 Hooray!!! Thank you so much for your support! :)
just wooow...never thought someone could explain things in this way.. GREAT WORK
someone give this guy an elixir of eternal life for him never stops to do videos like that
Thanks! :)
How can you understand what the heck he is talking about. It's like extreme amounts of math equations, can you solve those equations?
@@yashaswikulshreshtha1588 what is wrong with you? The math presented is NOT difficult, it is literally basic calculus. Yes, we do understand the math. If you are struggling with this, I recommend that you retake calculus and algebra then come back here once you have a good foundation. None of this will make sense if your math knowledge is weak.
@@navjotsingh2251 Yeah honestly sometimes I like maths but idk if it's me or my teacher never explained to me properly.
@@navjotsingh2251 Exactly when I posted this comment honestly I didn't have any of those courses neither calculus. So it was overwhelm for me
I am a professor in computer science, and did my dissertation on Bayesian methods in AI, and I will say that this is the best explanation of neural networks I have ever seen. It does no hand-waving, and makes no assumptions about the viewer's math skills. Coming late to the CS party in my 20s, and not being from a strong math background, I can really appreciate this approach. I will be using this video series in my Intelligent Systems class. Thank you.
Wow! Thank you very much! Since you like the video, I'll be shameless and say that you might also like...the book! I cover this and a bunch of other ML topics in The StatQuest Illustrated Guide to Machine Learning here: www.amazon.com/dp/B0BLM4TLPY
@@statquest I will check it out!!
*As a UA-camr myself, I really appreciate the amount of work that must have been put into this. Amazing video! :)*
Thank you very much! :)
All the best for your channel ..
@@abhishekm4996 Thank you! :)
100% Agree
As a non-UA-camr, I can also appreciate the amount of work that has gone into this！ ❣️
The tutorials are amazing. Love the simplicity you bring to the allegedly complicated concepts.
WOW! Thank you so much for supporting StatQuest! BAM! :)
For Christ sake, this guy is brilliant! Taking something very intimidating and showing how it works is worth gold. I am embarrassed to say that I am a scientist, have been for over 20 years. We over complicate and write papers to sound smart, but really never really say anything. And good luck ever trying to reproduce those results in the paper. This is how technology and science should be taught and discussed with our peers.
Wow, thank you!
People like you are shaping the future of students and people who are willing to learn. May God bless you and fulfill all your dreams. Josh you are such an amazing human being.
Thank you very much! :)
Thank you very much! :)
Triple BAM congratulations! I bled through this six years ago , and that took me months. Here, you condensed the concepts into 18 mins! and I still learned alot ! You are a blessing!! please keep posting more videos!
Wow, thank you!
I just can't express how grateful I am for these videos, you have a stunning ability to make concepts stick to my and other people's mind. You're the greatest in this field!
Wow, thank you!
You know you are contributing massively to the society when I look forward to your video releases just as much as I look forward to some of Marvel's blockbuster films!! :D
BAM! :)
@Jody Jones BAM! :)
Can I just say statquest videos >>>>> marvel films
@@unknownpleasures100 DOUBLE BAM! :)
These videos are just incredible. I can't imagine the effort behind each video. I'm just starting out in this and don't doubt that your channel will be one of my first options whenever I have any doubts. The simplicity and especially the way you explain it, is totally a 10. Thank you very much.
Wow, thank you!
Actual you are one of the greatest teachers of all times !
you make the most complicated subjects look sooooo easy that I was having breakfast while watching this ... and though I understood the dreaded Neural Networks 🤣🤣
Thanks alot josh, I wish you the best ❤
Wow, thank you!
Man, thank you for this. This is the most simple and concise way to explain NN I've ever seen online.
Glad you liked it!
The best Neural Networks explanation i've ever seen!! I went through many and all of them were pretty hard for me to understand. Keep up the good work man, you're doing great.
Wow, thanks!
I had tried to learn Neural Networks multiple times only to find that it gets more and more complicated due to the equations and terminologies. This is the first explanation I have come across that is so clear and concise and to the point, without going into equations, jargons. Thank you so much.
Thank you! I'm glad to hear that the video was helpful!
Truly truly truly appreciate your works, I introduce your works with my peers, and also my Profs, a million thanks. Watching your videos is the first step I do whenever I learn any new method.
Wow, thank you! And thank you for sharing! :)
We just enjoy learning and playing with knowledge with Josh. This is all because of his efforts, we're learning so well. Hats off to this man, spending his time spreading knowledge :D
Thank you very much! :)
These videos have explained the ideas so clearly in such a simple way. They literally helped me pass my actuarial exams. A big thank you for your time and amazing work!!
Wow!! Congratulations!!! TRIPLE BAM! :)
Been waiting for this topic ever since I discovered your channel a year ago. Can't wait to see the next parts!!!! Thank you, an amazing job as always :D
Thank you very much! :)
StatQuest with Josh Starmer me too,long for this topic
math, graphics, and progression shown combining the two is priceless. i couldnt get this from college professors, this video makes it clear. ...and i see how this now connects to Maths, maxima,minima,calculaus.... Appreciate the amount of efforts to make this visually motivating video.
Thank you!
Best series on machine learning topics I have ever seen. Clearly and precisely explained. Just awesome!
Wow, thanks!
Taking machine learning right now and my professor sucks, these videos are a huge help and help things make sense
I'm glad they are helpful! :)
Oh my, you just exaplained neural networks in a way that a kid would understand. Amazing!
Thanks!
now that was clarifying. And there are more parts in this playlist :) The amount of work you've put in these tutorials alone is worth a subscription. Thanks a lot
Thank you!
possibly the best explanation of neural networks I've ever seen. You condensed the concept into simplicity so effortlessly ! Thanks a lot Josh sir!
Thank you!
This is the best explanation of neural networks that I've ever seen. Thank you!
Wow, thank you!
Honestly I was so filled with joy when the notification came. I really missed your explanation in this topic. Thank you so much!!
@@ritikyadav157 Yes, I follow that channel too. His animations are also awesome!😍 But that time I couldn't understand it well because I didn't understand the previous things that is required to understand that. That is where Statquest came as a Savior!❤️
Thank you very much! :)
@@statquest Sir one heart pls🥺
@@GeekyMan lol!
@@hugolousa what's so funny?
Just brilliant. Seriously this explanation has turned on a switch. Thank you so much. This is the only video (out of many many I have watched) that explains the actual purpose of each node in the network and how they are combined.
I'm glad you the video was helpful! :)
When somebody says "Learning is fun" these videos seriously are the best proof. In lecture ive always felt overwhelmed by the maths and watching your videos breaks it down in such a simplicity that you have the necessary foundation to understand the complexity that follows. Thank you so much and happy new year 2024!!!
Thank you!
Your uploads are so timely for me! When you uploaded the XGB video, I was studying XGB. Now I'm studying neural networks! Thanks for this josh!
Double BAM! :)
The same for me .. and I was surprised not to find a video of neural network on your channel. Thank you statsquest
@@statquest Triple BAM! :)
same!!
I always had some doubts regarding mathematics behind neural networks, This video helped me so much regarding those! Next tier content!!
Glad it helped!
unbelievably helpful. your like the teacher that sticks with me past the point where i think im incapable of learning something but you help me unlock that lightbulb moment
Happy to help!
Fantastic! I have been reading and watching many articles and videos about Neural Network and this one is the most clear and well explained!! Congrats!
Wow! Thank you very much! :)
The best explanation for deep learning ever, can't wait for the part 2 video. Thank you for your wonderful job =)
Thank you! 😃
Love the humor! The calculation narration part was epic. Learned a lot!
bam! :)
Just awesome!!! Leaving out all the jorgons and learn only the underlying concepts, no need to look further just StatQuest :) I have become your fan, how well you explain such a complicated concepts in a easy and very neat format.
Thank you! :)
Dear StatQuest, we are expecting more videos on deep learning. Awesome.e explanation and intuitive thoughts behind each step. Kudos!
I will try my best
Perfect timing for me, I was just starting with neural networks. DOUBLE BAM!!!
Awesome! :)
I think the novel value of this video (for me) is that it helps bridge the gap left with 3Blue1Brown's video where he intuites what the NN may be doing, but then reveals that it actually appears random and chaotic, without the sense of order we initially assumed to make sense of it. This video explains that although the particular weights and biases may appear random at first glance, it is through their summation/cancellation that an order emerges. Also, kudos for introducing a new beginner example -- I genuinely did not realise that there were other applications outside of image recognition lol (I am a beginner to this!). As in, yes it's easy for us to look at the simple data and say 'hurr durr why not just model it with a quadratic equation'; but here is a different way to derive values to fit the data on the graph. Which was new to me, so, thanks!
BAM! I'm glad you liked the video and the example. :)
I was waiting for this video for a long time. Surely, This is going to be an amazing series.
Thank you! :)
My friend recommended this channel. You are so brilliant! This singing was a perfect touch! I am just learning these concepts and all the professors. Assume we are supposed to understand what’s going on from get-go even if it’s the concept I encounter first time in my life! Thank you!
Happy to help!
What a perfect time of uploading a perfect topic. Thanks a lot josh
Thank you! :)
I love your videos, thank you so much for getting us non experts closer to knowledge :)
Glad you like them!
I have done tons of coursework on NNs, worked with them for years but you taught me something I never even thought to ask. Amazing job Josh, well done and thank you!
Thank you!
These videos are so helpful. I’m happy I found your channel. INFINITE BAM 💥
Hooray! :)
That was great explanation. Hope you make videos more frequently and expand the content of NN to advance topics.
That's the plan! :)
When it comes to statquest videos, I will first hit the like button and then watch the video !!
Amazingly detailed explanation. Thanks a lot for making these videos !
Awesome, thank you!
had to promote this video by committing this is awesome and joining the membership. Will be a continuous supporter from now on. BAM!
Wow!!!! Thank you very much for your support! BAM! :)
Great energy, willingness to spread concepts and knowledge!
Thanks!
Man if only i had discovered this channel a while back when I was having my stats and data analysis classes, I wouldn't have performed so terribly x) Good thing that I found out about it now so I still can re-learn all what I missed on, now that I need that knowledge the most. Thank you so much Mr Starmer, you and your videos are wholesome beyond belief
Thank you very much! :)
your videos helped me through my statistics seminary last semester, the final project was about neural nets so I would have loved for this video to be available back then haha. now I'm writing my thesis on random forests and boosting. thank you for all your work
Hooray! Thank you. :)
Such a good video... I can't wait to finish the series! Thank you for making this. I love the way you explain everything, and your little jingles.
Thank you so much!
superbly explained, this man is just amazing.
Thank you!
Yes Josh!!! Finally on Neural Network! You're awesome!!
Hooray!!! Thanks!
I clicked on your video so fast once I saw you covered neural networks...it is so reassuring to know you will have all your answers in one place.
bam!
Outstanding. Thank you for the great combination of clear explanation, simple language, and tongue-in-cheek humor.
Glad you enjoyed it!
Amazing, Josh! I wish you and your channel existed in my life when I was younger! I'd definitely had done different career choices with this way of looking mathematics and statistics you presented in your videos! 🇧🇷🇧🇷🇧🇷🇧🇷
Muito obrigado! :)
Boa man
I have a quiz tomorrow, zoom classes suck. so thank you. You're a life saver
Good luck and let me know how it goes. :)
You have been spotted
I've been struggling to understand neural networks until i stumbled upon this video. This is the best explanation with the best presentation (I agree fully on using easy to understand visualization instead of those fancy one). I don't usually write comments, but I feel the need to thank you for this. Thank you so much!
Thank you very much! :)
hi..big fan from Malaysia.i got a 1 inch thick neural network's textbook that take a whole life to understand. n u just take a few minutes of my life to explain it very well. thanks a lot!
Great to hear!
That Helloooooooooooooo at the starting of the vid is so cheerful :3
Thanks! :)
Thank You so much for making this series, Love from India.❤️❤️
Thank you! :)
just started phd in astronomy and i know your channel is gonna be a pillar for my success. cheers mate!
Thanks and good luck with your PhD!
This is the greatest explanation I have ever seen in my whole life till now. The math + graphical explanation are too good.Thanks my brother for sharing. May God bless you. Please do keep sharing videos like this with us brother. Thanks🙏
Thank you very much! :)
@@statquest 🙏
I'm going to present my master's degree defence next month and this video, and the ones about PCA and SVM, helped me *a lot* to break down these complicated processes into something that I can actually try and explain to the professors that will be evaluating me. Thank you SO MUCH for that!
Awesome and good luck with your defense! Let me know how it goes. :)
@@statquest Thank you, I will! :D
@@statquest Just thought I'd let you know that I was approved and now have a MSc degree in biochemistry and molecular biology! Thank you very much for your help! 😃
@@lodjr TRIPLE BAM!!! Congratulations!!! That is awesome. Raise a glass for me when you celebrate! BAM!
@@statquest Thank you! I will! 😃
Please release a full album of your neural net calculation montage singing.
Can’t wait for the next video.
You made me laugh. Can you imagine a full 45 minutes of me beep booping? :)
@@statquest Here's hoping
To be honest I was eagerly waiting for you to upload contents on neural networks....can't wait to see your explanation of optimizers used in fitting neural networks.
Next up is backpropagation.
This wa so useful. Some concepts don’t make any sense if you read them out of a book and need a video. This was perfect.
Thanks you so much for the work you put in
Thank you!
Josh, that was simply fantastic. I've watched a lot of videos and taken a course in NNs, and I've never seen an explanation like this. This was exactly what I was looking for. I don't know how you know what you know, or why no one else seems to be able to teach like you do, but THANK YOU!!!
Awesome, thank you!
If I ever get a job after my graduation, I will donate money here... You are a GOD
bam! Good luck!
First time I saw someone creating curves from linear equations of weights and biases. I was searching for this for a long time. Thanks for explaining it so clearly.
Glad it was helpful!
Please don't stop posting videos like this. I'm planning to become a Machine learning engineer with help of your videos which is more than enough. What a man you are! oh my god! hat's off!!!!!!!!!!!!!
Thanks!
you literally save my life and human beings *crying inside*. Thank you indeed
Thanks!
Thanks Josh for helping remove the gatekeeping on Machine Learning for folks! It often is held as a this thing only prodigy's or math wizards can do, but you are helping to make it accessible for everyone! Keep up the amazing work.
Thank you very much! :)
I wish I could thumbs up your comment 10 times!
The usage of "gatekeeping" just gets dumber and dumber over time.
Thank you so much for this! I have been looking for something like this for a long time - the visualization really helps a lot to understand whats actually going on behind the hidden layer!
Glad it was helpful!
Good afternoon, Josh! I am another student who came here to express the abysmal gratitude for the explanations and work you put here. Thank you!!
Thank you very much! :)
This is the only video on the internet that has gone so deep into neural networks.
BAM! :)
@@statquest double that :)
I was waiting for this topic Sir, Thanks a lot.....
:)
WTH. That's all maths. I am quitting neural networks and ML. I guess gotta open a general store instead of being a AI researcher
@@yashaswikulshreshtha1588 bro if you don't like maths then may be you are not made for AI/ML( But if you work hard you can learn.... just need focus)
I cannot say it enough how much I feel grateful towards you for making these videos. God bless you
Wow, thank you!
I appreciate you effort to rectify misunderstandings, and more importantly, your generosity
Thank you!
You once fitted a line
Now you fit a squiggle
"Softplus" made me laugh
Self-promotion made me giggle
STATQUEESTT!
TRIPLE BAM!!! I love the poem. :)
@@statquest thankks!! it was actually a rendition of the logistic regression's "awesome song and introduction"😅
@@J10098 :)
The intro is what gives me joy. Everything feels so easy when Josh the God explains.
Thanks! :)
Josh I just wanted to say thank you very much for all this content, it's really enlightning and very powerful in a sense that just with simple and not so fancy explanations I'm able to nail down every single concept and idea. I'm currently taking a financial engineering masters degree and let me tell you that your approach has been really helping me out. Also, in form of gratitude, I bought the 'Everything Bundle' from gumroad and it's been quite an awesome experience to learn from it and sharpen some loose math and stat concepts that I had on me. Big thank you from Chile.
Thank you very much!!! And I really appreciate your support! TRIPLE BAM! :)
That's just amazing man. Your way of simplifying hard concepts is just out of the ordinary. Thanks for your efforts.
Thank you!
Just what I needed, epic
BAM! :)
Thanks again for yet another great series of videos Josh, I've covered the basics of neural networks in detail during my masters coursework and went through quite a number of resources (including the ones you've looked at which were recommended by my lecturer) to get somewhat of a grasp on it. Imho yet again for another key stat/ ml topic yours proves to be clearest and easiest to follow while being entertaining especially with the incremental 3 part video approach to backpropagation and step-by-step calculations at each step. (Which similarly proved invaluable in your works in gradient descent).
The series deserves so many more views, my gut feel is that these view numbers are going to go up when semesters get back in session? Have a great 2021 Josh and stay safe!
Thank you very much! I've got 4 more Neural Network videos coming out in the next month.
@@statquest awesome can't wait :)
I've never seen a clearer explanation than this. Thanks!! I think I'll watch the whole series now
Hooray!
15:49 - 16:18 -- someone's trying to break into the hyperpop/electro-pop scene with a brand new single :D
:)
No matter where you come from...you will definitely get something new,something interesting to learn from him...
BAM! :)
The best introduction to NN I ever saw! All the series is great, thank you
Thank you! :)
I'm studyng Machine Learning at Datacamp and several explanattions are very sheety. Your videos add so much light to my study path. Thank you!
Thanks!
Won't mind if there are more ads until i get such quality content. Please try to upload all DL videos so I can prepare for interview.
Thanks
I'm working as quickly as I can and hope to have part 2 out in a few weeks.
@@statquest Also please create a playlist section for them so we can find them easily. thank you!
@@deathkillertech Will do!
I was telling my friend today that StatQuest is bar far the best UA-cam Channel I have ever subscribed! 10 out of 10; the greatest job ever
Wow, thanks!
Whenever I go out with friends, in some way Starmer's name comes up hahaha. Thank you for this
BAM! :)
same!
@@saraaltamirano BAM! And thank you for your support!!! BAM! :)
Whenever I watch your videos, there is always a big, fancy, squiggly smile on my face. CURVED BAM!!!(thx a lot)
bam! :)
Till now you are the best one who explained the idea of neural network very clearly. I'm gonna continue watching the other parts as well, and of course buying some of your fancy T-shirts. THANK YOU SO MUCH.❣
Awesome, thank you!
Great Josh.
May Almighty Allah reward you.
Thank you!
Just a simple comment. Need heart asap.
BAM!
Thanks Josh. You help me in comprehending Neural networks more intuitively. It gave me the idea to work on my machine learning assignment! Kudos to you!
Glad to help!
I just love EVERYTHING about this video. Such a beautiful, intituitive, gracious explanation. Thanks!!
Thank you so much! :)