This repository contains lab exercises from an introductory class on machine learning and deep learning using CUDA, which I completed during my 4th semester. The course covered a wide range of foundational topics in machine learning and deep learning, with a particular emphasis on leveraging CUDA for accelerated computation.
-
Classification and Regression
- Types of problems that can be solved with machine learning
- Key differences and applications
-
Simple Classification Networks
- Importance of loss functions
- Activation functions
- Parameters like batch size, batch normalization, and optimizers
-
Introduction to Convolutional Networks (CNNs)
- Basics of convolutional layers
- Pooling layers
- Application of CNNs in image processing
-
Introduction to Residual Networks (ResNets)
- Concept of residual learning
- Architecture of ResNets
- Advantages of using ResNets for deep learning tasks
-
Introduction to Generative Adversarial Networks (GANs)
- Architecture of GANs
- Inner workings of the generator and discriminator
- Applications of GANs in generating realistic data