What is Back Propagation

Learn about watsonx→ ibm.biz/BdyEjK
Neural networks are great for predictive modeling - everything from stock trends to language translations. But what if the answer is wrong, how do they “learn” to do better? Martin Keen explains that during a process called backward propagation, the generated output is compared to the expected output, and then the error contributed by each neuron (or “node”) is examined. By adjusting the node’s weights and biases, error is reduced and thus the overall accuracy improved.
Get started for free on IBM Cloud → ibm.biz/sign-up-now
Subscribe to see more videos like this in the future → ibm.biz/subscribe-now

[@vencibushy]

Back propagation to neural networks is what negative feedback is to closed loop systems. The understanding come pretty much naturally to the people which studied automation and control engineering.
However - many articles tend to mix thing up. In this case back propagation and gradient descent. Back propagation is the process of passing the error back through the layers and using it to recalculate the weights. Gradient descent is the algorithm used for recalculation. There are other algorithms for recalculation of the weights.

[@Kiera9000]

thanks for getting me through my exams cause the script from my professor helps literally nothing in understanding deep learning. Cheers mate

[@anant1870]

Thanks for this Great explanation MARK 😃

[@hamidapremani6151]

Brilliantly simplified explanation for a fairly complex topic. Thanks, Martin!

[@hashemkadri3009]

marvin u mean, smh

[@Mary-ml5po]

I can't get enough of you brilliant videos. Thank you for making what it seemed to me before as complicated easy to understand . Could you please post a video about loss functions and gradient decent?

[@im-Anarchy]

What did he even taught actually?

[@sakshammishra9232]

Lovely man..... excellent videos..all complexities eliminated. thanks a lot 😊

[@ca1790]

The gradient is passed backward using the chain rule from calculus. The gradient is just a multivariable form of the derivative. It is an actual numerical quantity for each "atomic" part of the network; usually a neuron's weights and bias.

[@1955subraj]

Very well explained 🎉

[@neail5466]

Thank you for the information.
Could you please tell if the the BP is only available and applicable for Supervised models, as we have to have a pre computed result to compare against!!
Certainly, unsupervised models could also use this theoretically but does / could it effect in a positive way?
Additionally how the comparison actually performed?
Especially for the information that can't be quantised !

[@msatyabhaskarasrinivasacha5874]

Awesome.....awesome superb explanation sir

[@Zethuzzz]

Remember the chain rule that you learned in high school.Well that’s what is used in Backpropogation

[@rigbyb]

Great video! 😊

[@sweealamak628]

Thanks Mardnin!

[@idobleicher]

A great video!

[@rishidubey8745]

thanks marvin

[@guliyevshahriyar]

Thank you!

[@pleasethink4789]

Hi Marklin!
Thank you for such a great explanation.
(btw, I know your name is Martin. 😂 )

[@ashodapakian2788]

Off topic: what drawing board setup do these IBM videos use ?
it's really great.

[@boyyang1290]

I'd like to know, too.

[@boyyang1290]

I find it ，he is drawing on the Glass

[@jaffarbh]

Isn't Back Propagation used to lower the computation needed to adjust the weights? I understand that doing so in a "forward" fashion is much more expensive than in a "backward" fashion?

[@Ellikka1]

When doing the Loss Function hove is the "Correct" output given? Is it training data and the compared an other data file with desired outcomes? In the example of "Martin" how does the neural network get to know that your name was not Mark?

[@l_a_h797]

5:36 Actually, convergence is does not necessarily mean the network is able to do its task reliably. It just means that its reliability has reached a plateau. We hope that the plateau is high, i.e. that the network does a good job of predicting the right outputs. For many applications, NNs are currently able to reach a good level of performance. But in general, what is optimal is not always very good. For example, a network with just 1 layer of 2 nodes is not going to be successful at handwriting recognition, even if its model converges.

[@mateusz6190]

Hi, you seem to have good knowledge on this, can I ask you a question please. Do you know if neural networks will be good for recognizing handwritten math expressions? (digits, operators, variables, all elements seperated to be recognized individually). I need a program that would do that and I tried a neural network, it is good for images from dataset but terrible for stuff from outside the dataset. Would you have any tips? I would be really greatful

[@boeng9371]

In IBM we trust ✊😔

[@the1111011]

why you didn't explain how the network updates the weight

[@stefanfueger3487]

Wait ... the video is online for four hours ... and still no question how he manages to write mirrored?

[@Aegon1995]

There’s a separate video for that

[@itdataandprocessanalysis3202]

🤦‍♂

[@IBMTechnology]

Ha, that's so true. Here you go: ibm.biz/write-backwards

[@tianhanipah9783]

Just flip the video horizontally

[@mohslimani5716]

Thanks still I need to understand how technically does it happen

[@AnjaliSharma-dv5ke]

It’s done by calculating the derivatives of the y hats with respect to the weights, and the function done backwards in the network applying the chain rule of calculus

[@Justme-dk7vm]

ANY CHANCE TO GIVE 1000 LIKES ???😩