Deep Learning 8
- Weight Initialization of Neural Networks
- Bias-Variance Tradeoff and Regularization
- Model Smoothing to Prevent Zero-Probabilities in Probabilistic Language Models
- Negative Log Likelihood Explained
- Building a Bigram Character-Level Language Model
- Building a Multilayer Perceptron (MLP) from Scratch
- Linearity and Linear Transformation
- Derivative and Backpropagation