q5

Problem 2 - Gaussian Process Consider a parametic model governed by the parameter vector w together with a data set of input values x1,...,xN and a nonlinear feature mapping

Problem 3. Neutral Network Consider classifying a single hidden layer neural network for the

For this programming assignment you will implement the Lenet 5 CNN using either pytorch or tensorflow, but not Keras. You can take a look to other implementations in internet but please, when coding use your personal coding style and add references to your sources.

1. Design and develop a text classifier which can be used as an amazon review categorizer. Your classifier must be able to train to classify reviews into one of two classes. Positive and negative reviews. Description can be found in the readme file. Please note that we are using only the test set as the dataset is huge. This test set contains 400k data points. a. Data set can be found in the canvas b. Use the TfidfVectorizer found in Sciekit-learn library in python to vectorize the dataset c. Use GaussianNB for the classifier d. Calculate the accuracy of the model. You need to use the data partitioning to create train set and test set from the data set given. e. Input a sample text and determine the class of the text provided

3-For the data shown in the attached figure (dark circles are one class, white circles another) solve the classification problem with a neuron by hand. That is, find the appropriate weights of the required linear discriminant.

In this problem, we will use a linear binary classifier (no activation function) with weight matrix W of size M x F to illustrate the use of the gradient descent algorithm in updating W at each iteration. We choose F = 5 and M = 2 to represent the number of features (per one example) and number of labels considered, respectively.

2. (a) What is a linear scoring function? How can it be used for classifying test samples into positive and negative?[4 marks] (b) Consider the linear scoring function with parameters b = 2 and w = (2, −1, 1),where w are the weights. Calculate the predicted label for the test sample (c) What is the margin of a given separating hyperplane? (d) What is meant by the maximum margin hyperplane (also known as the optimal separating hyperplane)? (e) Define the notion of a support vector in the context of maximum margin classifiers.[3 marks] (f) Consider the following training set with two features: the positive samples are (0, 2) and (1, 2); ·the negative samples are (0,0), (0, 1), and (−1, 1). (g) Give an example of a training set for the problem of binary classification with only one feature where no separating hyperplane exists.[3 marks] (h) State an optimization problem whose solution is the maximum margin hyper-plane. Give the geometric interpretation of each formula in this optimization problem.[6 marks] (i) State the soft margin classifier as an optimization problem. Give the geometric interpretation of this problem.