Muhammad Umer sani : MSDS19063
Muhammad Talha Hashmi: MSDS19088
Mukarram Ahmad: MSDS19090
Agenda: Back Propagation.
Let’s us take an example.
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/1-1.png?w=347)
- Where w are vectors of same vector.
- Xn are inputs.
We can find back propagation by dot products.
Perceptron is something that you got by dot product.
σ (x)= 1/1+e^-2 => output σ (x^T.w)
We can do this for any computable form.
Question: What is bootable?
Answer: Which can be computable.
Question: What is example of bootable?
Answer: Above one is an example of bootable.
If domain and range is finite then it is computable otherwise its incomputable.
Lengthy program can be neural network.
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/2-1.png?w=185)
That is the tricky part.
All neural network are heuristic base. In the above example weights are missing, so we have to learn it.
We will start the optimization by minimization and maximization.
We will change neural network formulation of above equation.
That is (f’-f)^2
Question: How we can add more dimension?
Answer: We will discuss it later.
Question: Its neuron and perceptron?
Answer: No it is different thing.
As an example
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/3-3.png?w=141)
Getting output from above example is neuron
Question: Is there any mathematical form to adding neuron to over fit values?
Answer: No there is no mathematical form.
So lose function is
L = ( f’-f )^2 which we want to maximize
We will find the gradient is zero which is called a gradient descent.
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/4-1.png?w=288)
So dot product of weights and inputs
![Left Bracket:](https://itu18.wordpress.com/wp-content/uploads/2019/12/5.png?w=265)
As we have input and output so weights will be random
L (w,b = (a1^3 – y1)^2)
It’s a first last function for fixed w.
Lets find out gradient descent of this function.
δL / δw
δL / δb
we define δj^l = δL / Zi^l = δL / δaj^l , δai^l / δzi^l
We can compute this recursively
δ’ (x) = δ (x) (1- δ (x))
this function is for fix layer not for all layer So δ’ is error
Question: WB have predict value, so we also have actual value.
Answer: No, we do not have.
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/6.png?w=337)
Question: Why we are δ calling delta error.
Answer: So let’s we will call delta an error.
Again seeing the same example
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/7.png?w=188)
Taking derivative w.r.t ak^l as it will not effect the b so.
![](https://itu18.wordpress.com/wp-content/uploads/2019/12/8.png?w=299)
Which was our goal.
At the end two quiz’s was conducted and discuss the final syllabus, all the content of semester will be include in the exam and sir will provide the formulas sheet but every one should know the major formulas as well.
Best of luck for your exam..
is this included??
LikeLike
Yes
LikeLike
Did you mean to say: “Where w are *WEIGHTS of same vector” ?
LikeLike
Yes.
LikeLike