My First Course: Introduction to Deep Learning from HSE
Thanks to mine being a student at National University of Colombia, I’ve been granted access to a set of Coursera specializations. I decided to take the Advanced Machine Learning Specialization, since I want to be able to understand the basics of this filed, and how to apply it to the study of quantum systems.
At this stage, My knowledge of Machine learning is pretty basic. I haven’t worked with any python framework, nor have I went beyond simple linear fits in my lab classes.
In this blog, I will summarize my learning process and also some perspectives and exercises I might probably encounter in the journey. Today, I start with the first course: Introduction to Deep Learning from HSE.
Course Intro
This course has the following prerequisites:
- Machine Learning Basics
- Probability theory
- Linear Algebra
- Calculus
Most of them I have in decent amount, except for Machine Learning. The structure is as follows
| Week | Topics |
|---|---|
| 1 | Linear Models: Separating data classes using a plane |
| 2 | Multi-Layer Perceptron: Separating classes with non-linear boundaries |
| 3 | Convolutional Neural Networks (CNNs) and Image processing |
| 4 | Working with neural representations: Encoding images |
| 5 | Recurrent Neural Networks and text |
| 6 | Training an image captioning model |
Basically, since model training requires a huge amount of computational resources, I will use Google Colab for all the assignments and exercises. At least while I can get my hands on a PC with a powerful enough GPU.