Deep Learning

This section explores neural network architectures, training dynamics, loss landscapes, and optimization algorithms from first principles.

Key Focus Areas

• Core Architecture: Multi-Layer Perceptrons (MLPs), backpropagation, activation functions.
• Optimization & Regularization: Gradient descent variants, batch normalization, dropout, weight decay.
• Advanced Topologies: Recurrent Neural Networks (RNNs), LSTMs, Transformers, and modern generative architectures.