A Brief Introduction to Neural Networks (unfinished series)
Вставка
- Опубліковано 3 тра 2018
- This series is intended as a light introduction to neural networks, with a focus on the task of classifying handwritten digits. All programming will be done in python (v3.6).
Support the creation of more tutorials:
/ sebastianlague
Hi everyone, sorry for leaving the series unfinished. I do however have a new video out now which covers everything I intended to talk about in this series (although the code is in C# instead of python). You can watch it here if you’re interested:
ua-cam.com/video/hfMk-kjRv4c/v-deo.html
It is amazing that I have been waiting for 5 years for this and end up being 9 months late to the update.
Regardless thank you for the amazing discussion and I look forward to a few extras you could give us about other methods for speeding up the networks self growing networks maybe or any other good ideas
This is by far the best visual example I have ever seen on how the weights and biases change the final output. Great work! Looking forward to the rest of the series.
Cap n CoCo yes same
Agreed. Excellent visualisation!
I wanted to put the same comment dude! Awesome!
Thanks!
Because Sebastian is a wiz on the PC. I remember I had to watch lecturers trying to explain it with a cluttered white board and different color markers. This here makes so much more sense and is much clearer. Thanks Sebastian, looking forward to the rest of this series.
Seriously. I echo everyone else's comments, This is by far the best video I've seen on this topic. Outstanding work.
And one that explans it through decision boundaries, which is pretty rare
Honestly, I agree this is so good
So true!! Perfect!
I have watched maybe around 50 hours of footage trying to explain topics of neural networks over the years. This is the best one yet.
I'm reading a book which explains ANN's. After four chapters on it I came to the conclusion that I'm not going to get it. Just watched your video by chance and now all I read makes sense. Great job!
same with me
excuse my french but holy FUCK this is well done. The way you skip unessential bits so that the viewer can grasp the whole before going in to details is great. I personally prefer this mode of education, even when I'm learning seriously, it becomes much faster to learn the details after understanding the goal
I mean, 3B1B's videos on neural networks are great, and they have some great explanations, but in this video, you explained the "network" part of a neural network in just the right way... in five minutes!!
This is strange i have been taking the machine learning course on Coursera for the past 3 weeks and also checking around online and You by far have the best description of what is going on. You basically get the much faster than anybody has. This is really really strange.
The algebra/calculus and decision boundary. I mean I feel like i should stop the course and wait for you
I took that Coursera course a couple of years ago and I found it completely underwhelming compared to some resources that are available for free on the internet.
It's his voice and the trust he builds by showing you a working example...so you listen and try to understand what he is actually saying.
If a person were to say the exact same thing in a voice with variations...your concentration would break, even though the video is the same
@@originaltenka5545 fuck good point
@@originaltenka5545 that's a good point, studies show people trust/feel soothed by someone speaking with a British accent
Oh funny brains!
Anna Med I can even tell someone is British just by the fact that I trust him
I have never seen this plotted as a graph before - it's all suddenly very simple! Great video
Wow... I never knew why the activation function was necessary, but only knew it needed to be there. Now I know! Thanks so much for making this very easy to understand!
Not only that, but the activation function needs to be non-linear in order for the neural network to create its own intermediate features. Any non-linear activation function will do... even ReLU
After having neural networks at my university - this 5 minute video enlighted me with this visualization. This is pure gold
WOW! this is genius! I'm mean the way you explain it is incredible: very straightforward but nonetheless amazingly efficient.
This is a stunningly clear explanation of how neural networks work, in just under 6 minutes! Amazing. You have exceptional clarity of thought, and it shows in how you explain things. Thank you.
This is the best video i found regarding the neural networks, i'm not even exaggerating.
This has to be the most intuitive explanation of a NN, It actually explains why a neural network does what it does in the most simplest terms. Even beats the video by 3b1b if you ask me! Well done!
I have watched so many NN teaching videos but this is by far the most helpful and easy to understand video on Neural Network. Finally I understand how the weights and the Bias work.
Thank you so much.
That was an incredible way of explaining WHY features of a neural network are there, especially the intuition given for adding nonlinearity - something I've only seen most people gloss over! Awesome work!
Wow, just wow. I've never seen such a fantastic tutorial. I'm blown away by how much easier this is to nibble on as opposed to other tutorials out there.
Thank you so much for making this video!
You're a freaking mad genius. This is the best neural network video I've seen.
i never subscribed to anyone this faster!
been self studying ML and NN for over 4 months and only now i have a clearer understanding of why biases and activation functions (except ReLU) are used !
this video is gold
This is BY FAR the clearest, most intuitive explanation of neural networks I've ever seen. Well done!
This is probably the best explanation I’ve ever seen of a neural network’s inner workings, and in only a few minutes.
Best, most intuitive/visual explanation I have seen about how neural networks actually work. Now I actually understand what the activation function is for and I have a better understanding of neural networks in general
It's possibly the best description I've seen so far, but again, I find myself at that point where I ALMOST understand it.
The thing I can't really grasp is what exactly the hidden layer does, or how to determine how many nodes it requires.
I have had some success with some classifications, where I have just randomly changed the number of neurons in the hidden layers, or number of hidden layers, or a combination of both. But without understanding what it does or how to know how many neurons I will need, it's just magic.
And Sebastian here did the same as every other tutor I've come across so far. "And then we'll add a hidden layer with three neurons.." and that's it. That's the explanation we get.
I saw another video, on classification of handwritten numbers, where it was explained that one layer would preferably check for patterns in the writing and the next would check for patterns within those patterns, but no explanation to how that worked when both had the same calculations and the same activation function, nor any explanation on how they came up with that exact number of neurons in each hidden layer.
And when visualizing it, he determined that it didn't check for those patterns at all, yet it did classify the numbers correctly.
So again, without understanding it, it's just down to "magic".
Best visual explanation so far in internet, that's what I need.
Best video explaining the topic, finally a video that doesn’t assume you have a degree in maths with some great visualisations
Seeing this in a visual aspect, and seeing the activation function cause non-linearity really helped everything click in my head. From 5 minutes I've gained a large amount of understanding I hadn't gotten elsewhere before. So thank you for that!
This is by far the best visual explanation I have ever seen explaining NN.
I am impressed how well it is explained and the animations illustrate perfectly what you are explaining.
As a practitioner, I never had the privilege of being able to explain neural networks with so much ease to audiences who have little or no idea on how activation functions, biases and hidden layers work. You, my friend, are on the top of my list - a big thank you! I look forward to more content like this, keep up the excellent work!
This is the best illustration of a perceptron model I saw ever!
This has just made this topic 100000% easier to understand thank you so much!
This is the best resource I have seen while trying to understand neural networks. The visual representation helps a lot...
I love this video so much. The way you explained it so simply and using pure linear equations, it's beautiful. Thank you so much for this video.
never saw a better explanation of this, even understanding it before it was super cool to see it adjust by hand sliders and how activation function changes it
Man, your channel is a treasure. You explain complicated stuff in the easiest and quickest way possible. Best regards.
I'm actually mad that after reading actual books on the subject, this 5 minute video was the most intuitive representation of neural networks.
OMG! This is the best visualization I have seen to date to show how weights and biases actually manipulate the outputs.
Honestly the best explanation I have seen so far. Well done.
Only just discovered your channel. It's gold, the explanations are simple enough you can understand it without any other context, but you provide enough detail it's possible I could write code based on what you said.
Best video on Neural networks I have ever come across, very crisp and clear.
i have genuinely learnt more in 5 minutes watching this than i have searching the internet for a week. Cant wait for the rest of the series.
Being a Deep learning researcher and having taken courses both in college and online MOOCs I find this as the best visual explanation of neural networks ever. Huge fan of your channel, been watching it for unity for a while now, and will now start watching your channel for everything. Keep up the great work Sebastian.
Best explanation of NN I’ve encountered and the first time I’ve seen it visualised.
2 years late and I barely comment but this is by far the best video on neural networks I have ever discovered.
After more than a month of searching, clearest video I've seen
Seeing everyone comment on how well you explained this I realized I am not the only one thinking you did an amazing job!
I have been searching for a good example of neural net and this is by far the best one. Thank you, Sir!
Best explanation ever. So simple yet understandable
This deserves much more attention. Great job!
I’ve watched about a dozen neural network videos. I have taken Calculus in college and the math does not intimidate me. However you have shown how the weights and bias and activation function actively change the output function and it finally clicked in my brain what was going on. THANK YOU.
WOW! I have never seen neural networks explained like this! So intuitive and creative. Thanks
Truly amazing video... Magnificent presentation, beautiful code work. Thank you alot for all of this!
After watching so many videos about neutral networks, I really didn't think that any other way of understanding this thing can exist. But Sebastian, you've done it, again. You have a real talent there...
I've seen so many different explanations of this basic neural network over the last 12 years, and this is by far the best! Your step-by-step build up of the visuals along with the network was excellent!
These are probably questions too involved for such a light introduction, but... How do you determine how many nodes will be needed within the hidden layer to get the result you want? Do you have a deeper explanation in one of your other videos as to how the hidden layer helps and why it's needed? I thought it was interesting to see that adding the hidden layer didn't help (immediately), but then adding the sigmoid along with the hidden layer gave the desired behavior. I guess, I'm curious what would have happened with only the sigmoid function and no hidden layer.
I already viewed many videos on this subject, but this one is the clearest. Simple example, explaining draws and step by step learning. As usual, awesome work, thanx !
I think I learnt more in this video then in the 3 weeks of a degree level course on it.
You've got a way of teaching that helps given an intuitive understanding of the topic;
especially with the graphics!
Thanks for the video!
Sr, you are the first one that I found to explain this, you revived my interest to learn neural networks and IA, thank you
I would have not imagine I could find a better explanation than the one from 3b1b, but here we are!
Hey Sebastian! I got a fulltime software developing job because of your tutorials! I'm so appreciative to you because when I first watched your tutorial I was just a beginner in C#. What a Quality tutorials!
This is THE best explanation of a neural network!!! I’ve watched a bunch of videos, and this one takes the cake!!
Awesome, been looking forward to something like this from you. Can't wait to see more of this
Will do, thanks! :)
Excellent video with a clear and simple explanation of how neural networks operate. Thanks for this.
This is the most comprehensible video about neural networks all over the internet!
Incredible description, succinct, simple and quick!
Incredible work. Never seen such a clear explanation of this topic. Thank you.
Amazing as always, can't wait for the rest !
Excellent video. I have learned more in a few minutes here than I have learned from hours of watching other videos on the topic. The visuals really crystallized the principles in my mind. I could never figure out what the biases were for and I did not appreciate the full effect of the sigmoid function.
Eagerly waiting for the whole series! Thanks for the explanation. Really helpful.
Even if it will be stating what all the others have said before, this is simply the best explanation I have seen on the net and believe me I have checked a lot of them!
OMG, this tutorial insults MIT one hour lecture on neural networks . You are amazing Sebastian. So excited for this series.
An amazingly simple explanation for a not so simple topic.
You can't find this level of quality in paid courses even. Massive respect.
Holy heck this is the best explanation of neural networks I’ve ever seen.
Awesome! I'm really excited about this series. Nobody can explain stuff better than you can! Keep up the good work
This is my new favorite channel. Thanks !
my uni didn't come anywhere near explaining it this well. Thank you!!
Done thanks
1:30 in neural network with 2 inputs and 2 outputs, the decision boundaries are areas on the 2D graph (input1, input2) where the value of one output node is greater than the value of the other output node.
I am so happy, it has been so long that I haven't found your videos I just assumed you stopped them
This has totally cleared up how a neural network works! Thank you so much! Can’t wait for your next video.
I did neural networks in college and even had a book on them. I understood the general concept but never had an intuition for it.
Now I do, in 5 minutes. And I'm confident I can show this to almost anyone and they'll get it.
Yes I know there's another 300 comments saying this already but I think you deserve another one. Well done, good job, have a +1, a like, a gold star and a virtual beer
Best explaination of NN I've seen and heard so far. Thank you!
This makes it much easier to visualize how the networks learn. Like it could learn the difference between apples and oranges using a graph that is roundness in one direction and color in the other. Thanks!
This is really the best explaination i've ever seen
It's amazing how well done are the animations you use in your videos to explain how "things" work. There must be a lot of work behind them! Thank you!
Fantastic. Nice balance of theory and practicality.
Can't wait to watch the next episode!
I've only come across this now but oh my god this is amazing! I've always wanted to learn more about neural networks, but couldn't find any videos past "these neurons are connected, and these connections do stuff"
I'm super excited to see the rest of this series and hopefully start trying to code some of this stuff!
this is the best explanation I've ever seen
I've been trying to figure out neural networks for months now thank you so much
GREAT STUFF ! by far best explanation, I have heard. I have listened and read dozens. I never like to proceed until I thoroughly understand the step I am in . I wish I could give 5 thumbs up.
Yeah, this is by far the best visual example of some of the basic math and it's effects in NNs. Great job, I will be watching the rest of the series and have subbed!
I think you have made the greatest explanation video for neural networks on UA-cam ! Can't wait for the next videos in this series
I can't wait to see episode 2!
What a wonderfully illuminating illustration of neural networks. I’m genuinely impressed and grateful. You’ve earned a sub for sure.
I'm so glad I came across your channel! Your videos are super informative and well structured. I'm learning tons! Thank you
I love the way you explain these things. It really sinks in :)
Oh Yes!...You inject my brain correctly within this OnPoint 5+ minutes visual explanations, Bravo! Keep it up! 🤘👏
Definitely one of the best videos about neural networks (what a coincidence, I was just looking at your blender tutorials but how to apply neural networks is my second goal atm)
clearest visual explanation so far