Machine Learning 2021-II - G1

Dear student, please answer the questions of the following survey:

Answer the survey here

Topics

• Introduction

• Supervised Learning

• Unsupervised Learning

Link for meeting (or virtual lectures) - Every Thursday at 2 PM

Google meet: https://meet.google.com/aug-yvzc-znk

Video recorded lectures (You can watch them as long as you have an institutional email account, otherwise you must ask permission):

• Lecture 1 (2021-10-14)

• Lecture 2 (2021-10-21)

• Lecture 3 (2021-10-28)

• Lecture 4 (2021-11-03) (part 1/2)

• Lecture 4 (2021-11-03) (part 2/2)

• Lecture 5 (2021-11-11)

• Lecture 6 (2021-11-17/18)

• Lecture 7 (2021-12-02)

• Lecture 8 (2021-12-09)

• Lecture 9 (2021-12-16)

• Lecture 10 (2022-01-27)

• Another version of lecture 10

• Lecture 11 (2022-02-03): parts 1/3, 2/3, and 3/3.

• Lecture 12 (2022-02-10)

• Lecture 13 (2022-02-16): parts 1/3, 2/3, and 3/3.

• Lecture 14 (2022-02-23/24): parts 1/2 and 2/2

• 
  

Data sets 'n' source code:

• Data set used in Lecture 4

• Complete the following notebook according to what you watched in Lecture 4: Assignment 01

• Download the LaTex template for writing reports by clicking here! Besides, you ought to download the following file by clicking here!

• Don't forget to become a Kaggler and use the source code from the first assignment (see above the Assignment 01) on the following dataset. This is a quite known dataset, whose values of each record variables are a flower's features (e.g., petal width and length) and its species. In this new assignment, the goal is training a single artificial neuron to classify a flower as Iris-Setosa or not despite there are another two classes (i.e., Iris-Versicolor and Iris-Virginica). For further instruction you ought to check out the Lecture 6.

• Dataset for regression

• Assignment (polynomial mapping and regularized ridge regression)

• Last Assignment: Deadline is due 25th February at midnight

Grades:

• Grades of the first course part

• Grades of last course part

Resources:

• Deep learning book by Goodfellow et al.

• Pattern Recognition and Machine Learning by Bishop

• Elements of Statistical Learning by Hastie et al.

• Information Theory, Inference, and Learning Algorithms by MacKay

• Numpy website

• Scikit-Learn website

• Pytorch website

• Colab website

• Kaggle website. If you want to learn to harness machine learning, you ought to become a kaggler.

• UC Irvine Machine Learning Repository. If you need a dataset to work with, this is the right place to start your searching

• CRISP-DM for Data Science

• Handwritten training set at Hastie's webpage. Normalized handwritten digits, automatically scanned from envelopes by the U.S. Postal Service.

• Handwritten test set at Hastie's webpage

• MNIST database