#2. Solved Example Back Propagation Algorithm Multi-Layer Perceptron Network by Dr. Mahesh Huddar

#2. Solved Example Back Propagation Algorithm Multi-Layer Perceptron Network Machine Learning by Dr. Mahesh Huddar
Back Propagation Algorithm: • Back Propagation Algor...
Derivation of Back Propagation Algorithm: • Derivation of Back Pro...
#1 Solved Example Back Propagation Algorithm: • #1 Solved Example Back...
#2 Solved Example Back Propagation Algorithm: • #2. Solved Example Bac...
#3 Solved Example Back Propagation Algorithm: • #3. Backpropagation So...
#4 Solved Example Back Propagation Algorithm: • Backpropagation Solved...
Back Propagation Algorithm with bipolar weights: • 16. Update weights usi...
Multi-Layer Perceptron LearningSolved Example: • Solved Example Multi-L...
Multi-Layer Perceptron Learning: • Multi-Layer Perceptron...
Gradient Descent Algorithm: • 2. Gradient Descent Al...
The following concepts are discussed:
______________________________
Solved Example Back Propagation Algorithm,
Back Propagation Algorithm Solved Example,
Back Propagation Algorithm,
Multi-Layer Perceptron Network,
Back Propagation Algorithm Machine Learning,
Back Propagation Algorithm Multi-Layer Perceptron Network
Derivation of Backpropagation algorithm: • Derivation of Back Pro...
Gradient Descent Algorithm: • 2. Gradient Descent Al...
Gradient Descent and Delta Rule: • 1. Gradient Descent | ...
Machine Learning - • Machine Learning
Big Data Analysis - • Big Data Analytics
Data Science and Machine Learning - Machine Learning - • Machine Learning
Python Tutorial - • Python Application Pro...
********************************
1. Blog / Website: www.vtupulse.com/
2. Like Facebook Page: / vtupulse
3. Follow us on Instagram: / vtupulse
4. Like, Share, Subscribe, and Don't forget to press the bell ICON for regular updates

[@PritamKumar-mr5dv]

bias(thetha) calculated =previous_theta +learning rate*delta

[@amalayakin4411]

your equation is correct thank u bro .

[@muhammadharis7318]

how to calculate delta ?

[@xenon_429]

@@muhammadharis7318 we have calculated it before

[@chiranjeevikumarbattula8709]

For example we calculate theta(6)=previous theta (6)+learning rate*delta(6)

[@alexmanuelsommer]

Internet says it is bias(new)=bias(old) - learningrate × delta

[@MenchieExtrakt]

This video helped me out so much. I was so confused by my class meterials, this really cleared things up for me. Also, maybe remind people that beta is x0 and w0

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@tejasvinnarayan2887]

Thank you. Could you please give a clear guide on how to update the bias?

[@usmanalyassirajuddeen]

delta bias i = n(learning rate) * Sj(Error Term)
bias i (new) = delta bias i + bias i (old)

[@HiteshDewangan-y9i]

Sir how to find the weight for bais term? i am unable to understand

[@paulocorreia4694]

Hi Mahesh, thank you for your videos. They are super useful.

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@GabrielCephasIko-ojo]

Thank you so much for your lessons, they've been really helpful.
Clarification: Why is it that in calculating Error, you are not taking the square and dividing by 2
E=1/2(t-y)^2
I look forward to getting your response.
Many thanks

[@DPELHYSalmanFaizi]

How To calculate theta 6???

[@sanjeevareddyn]

Your explaination is crystal clear and well simplified. Please make more such videos on deep learning using different algorithms.(ReLU etc.,)

[@MaheshHuddar]

Ok
Do like share and subscribe

[@d-pain4844]

Sir in dervatiom video you said error is
Error = 1/2 £(td-od)²
Here you selected (target - actual) with no square and not even divided by 2??

[@jetnetgaming3594]

It's the derivative

[@pakfood9973]

If I use this method not derivation method both have same result?

[@hajrawaheed9636]

@@jetnetgaming3594 but the derivative should be used during backprop only, when in forward pass if we are calculating error we should use the formula as it is. anyways i think in the video they have demonstrated the error for the backprop purpose only hence they have calculated the derivative value. Can u please also confirm if these formulas and the chain rule derivation are the same?

[@prajwalm.s7976]

How to update bias values

[@prometheusthesis]

how do you update the bias value ? which formula ?

[@ToriliaShine]

i don't believe the bias value changes, only its weight. to change the bias weight, you use the same formula used to change the input/hidden nodes weight

[@ToriliaShine]

new bias weight(0,6) = (previous bias weight) + (learningRate*delta6*biasValue)

[@kcmokhk]

Thanks for lecture. May I know how the formula will change if I use ReLu activation function?

[@edgarpachisnky2946]

Thanks for tutorial , helped me a lot

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@samaysingh9978]

how bias are updated

[@nilabalasubramanian594]

For bias updation,
learning rate * y6
Add this with old bias

[@shadeslayer07]

Eta * delta for particular node

[@dr.ranjinips3103]

Thank you so much sir. Its really good

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@spokenenglishmentsbd9998]

love from Bangladesh

[@farheenshah-o2p]

how did you calculate the value of e ????

[@kavindasandalanka1056]

thank you very much. this video is very helpful

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@dhanushvcse4056]

Sir when i how to stop a error

[@nobuildwelp]

Thankyou mahesh bhai kaash aap hamaray teacher hotay abdullah kae abu hotay

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@dr.satishchinchorkar3872]

Thank you for your very crisp and clear explanation.

[@MaheshHuddar]

Welcome
Do like share and subscribe

[@AhmedImtiaz-s6g]

great video

[@MaheshHuddar]

Thank You
Do like share and subscribe

[@atharvafreefire7230]

updation of bias given in:
vid-83
ruclips.net/video/tTjcakAuHPI/видео.htmlsi=zvK_x5-JylavBZOz
(new)bias = (old)bias + Δbias
Δbias = learning rate*δj
δj is the error term Oj*(1-Oj)*(Tj-Oj)

[@the_candid_mechanic]

thank u so much

[@MaheshHuddar]

👍

[@bhavinarang4919]

How to update the bias, Sir?

[@sanjaykumar-ki9hv]

Use same formula for as we used in calculation of changing whieghts but take input weight of bias as 1 so final formula will be ⌂ = n * sj ,after that calculate new bias by adding old bias and the bias we calculated.

[@luthermillamuculadosreisec3844]

Thanks for saving me

[@MaheshHuddar]

welcome
Do like share and subscribe

[@SourabhKumar-nr1yq]

🙏🙏🙏🙏

[@MaheshHuddar]

Do like share and subscribe

[@danielwaweru7423]

Thanks.

[@MaheshHuddar]

Thank You
Do like share and subscribe

[@carloscosta4713]

I am brazilian.
Parabéns pelo excelente trabalho. Ganhou mais um inscrito!!!

[@MaheshHuddar]

Thank You

[@hwangdeokthien]

Complicated knowledge turns out to be very simple and easy to understand!! Thank you very much, sir!!
Hope you the best

[@MaheshHuddar]

Thank You
Do like share and subscribe