This course provides a robust curriculum on the foundational principles of Deep Learning (DL) and its essential role in the Generative AI (Gen AI) landscape. The learning journey starts by defining the path towards Gen AI, clearly answering key questions such as “What is Deep learning?” and “Why is it a foundation to Gen AI?” A core segment of the course focuses on the theoretical and practical difference between Machine Learning and Deep Learning.
A significant portion is dedicated to the mechanics of a neural network, detailing the processes of the forward pass and the crucial learning algorithm: backpropagation. Leaners will gain an in-depth understanding of parameters and the learning mechanism through practical understanding with the TensorFlow playground and a network simulator.
The curriculum then advances to critical applications, specifically sequence modeling. It explores the challenges of handling sequential data, justifying the need for specialized architectures like Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM). The course thoroughly explains the internal workings of RNN and LSTM, focusing on why these models require memory to effectively manage and process long sequences, providing students with a complete skill set in both foundational DL and advanced sequence processing. This is part-1 in the series. Next part will focus on Attention models, transformers and GPT.
