fa25-cs540 cs540
Lecture 12: Neural Networks II Lecture 12: Neural Networks II
Content locked
The catalog is public. Reading this content requires approved access.
Sign in to readLecture 12: Neural Networks II
Introduction
Multi-Layer Perceptron Review
The XOR Problem
MLP Architecture
Matrix Notation
Multi-Class Classification
Multiple Output Neurons
Softmax Function
Loss Functions
Binary Cross-Entropy Loss
Cross-Entropy Loss (Softmax Loss)
Calculus Review
Derivatives of Single-Variable Functions
Derivative Rules
Partial Derivatives
Gradient
Gradient Descent
Algorithm
Minibatch Stochastic Gradient Descent
Non-Convex Optimization
Computing Gradients with Backpropagation
Single Layer Example
Two Layer Example
Gradients with Respect to Inputs
Practice Problems
Key Takeaways
Suggested Reading