EIP2 : External Internship Program is a handson machine learning course being taught by The Inkers .
Concepts covered during Session 2
 Normalization and Batch Normalization
 Softmax activation
 ReLU and LeakyReLU
 Activation functions
 Gradient Descent and Stochastic Gradient Descent
 Backpropogation and its importance in ML
 Interesting new items heard during the Session
 Assignment Questions for Session 2
 Assignment answers
Interesting new items heard during the Session
 Hebbian Theory
 Fast.ai  MNIST training in 2.5 Hrs
 eNAS  Efficient Neural Architecture Search by Google : An ML which builds another ML
 3Blue1Brown youtube channel for Calculus
 HeartRate Detection Project by Interns at Inkers in colloboration with JohnHopkins University, Bill and Melinda Gates Foundation
Assignment Questions for Session 2
Write articles on the topics mentioned below between 50100 words in a markdown format .
 Convolution
 Filters/Kernels
 Epochs
 1x1 convolution
 3x3 convolution
 Feature Maps
 Feature Engineering (older computer vision concept)
Bonus points if you write on any of these:
 Activation Function
 How to create an account on GitHub and upload a sample project
 Receptive Field.
 10 examples of use of MathJax in Markdown
Assignment answers
Name : Sachin S Shetty
Batch : 4
 Convolution
It is the basic multiplication of matrices , ie. image matrix of size n x m into the kernel/filter matrix of size p X q . The result would be a convoluted matrix which would have either 1. increase the dimension of the image, for expanding the feature vectors 2. Decrease the dimension for analysis.

Filters/Kernels
The kernel/filters are one part of the matrix multiplication for convolution . These matrices are generated in random based for required distribution, ie. Gaussian. The randomness allows the variety of images/features to be clubbed together by backtracking.

Epochs
Epoch is the counter for running the entire convolution & backtracking operation with a reduced set of images. All the images from the training set are not taken completely in a single run, the images are taken in batches to train and improve the accuracy. In each epoch, multiple operations take place . The accuracy is calculated at end of each epoch and then experiment can either continue or else stop based on accuracy .

1x1 convolution
Reduces the dimension of the feature map after conolution operation .

3x3 convolution
Increase the dimensions of the feature maps after the convolution operation .

Feature Maps
Feature maps are the result of convolution operation, i.e the multiplication of feature matrix(images) with the filter/kernels.

Feature Engineering (older computer vision concept)
The earlier standard for solving computer vision problems before the advent of ML/AI, which in part was helped due to cloud computing was Feature Engineering.
Feature Engineering techniques like hough transform for detecting lines/circles, sobel filter are still prevalent for basic CV tasks.
These technique have the feature vectors/calculation hardcoded .ie. the values have been calculated and fixed to solve a particular problem/domain .
Bonus points if you write on any of these:

Activation Function
The activation functions helps to reduce the range of the matrix values/signal values. The prevailing standard function used in ML/AI is ReLu : Rectified Linear Unit, it rejects the negative values and passes 0 instead.
few other functions used earlier were sigmoid
 tanh

How to create an account on GitHub and upload a sample project. Link to have a personal website is sample_blog.

Receptive Field.

10 examples of use of MathJax in Markdown