Introduction to Neural Networks Syllabus

This course provides a (really) brief introduction to the foundational concepts of neural networks, their structure, functioning, and applications in machine learning.

Course Structure

• 1 lecture (3 hours): Slides
  • Overview of neural network basics: layers, weights, biases, and activation functions.
  • Key concepts in training, including forward pass, loss functions, and backpropagation.
  • Introduction to common architectures like Feedforward, CNNs, RNNs, and Transformers.
  • Optimization techniques, hyperparameter tuning, and regularization strategies.
• 1 session of exercises (3 hours): Statements
  • Calculations of forward and backward passes.
  • Analysis of scaling techniques for preprocessing.
  • Exploration of activation functions.
  • Gradient checking and its importance.
  • Regularization and its effects on model performance.
  • Critical analysis of common issues in neural network setups.