Notes.
9 notes
Why the median minimises absolute error and the mean minimises squared error, by nudging the guess.
A mathematical textbook connecting likelihood, entropy, KL divergence, statistical estimation, and probabilistic machine learning, with worked examples and solved exercises.
One hidden coin connects conditional independence, Bayesian updating, prediction, Naive Bayes, and maximum-likelihood estimation.
Understanding the kernel trick in SVMs, focusing on the roles of alpha and the inference formula.
The sigmoid and logit functions connect unbounded linear predictions to probabilities through log-odds.
Resources on regularization connect Lᵖ norms and their geometry to sparse model weights.
Adding a constant feature folds the intercept into linear regression’s weight vector without conflating it with statistical bias.
Driver risk assessment from dashcam footage, using a ResNet50 + LSTM model to classify driving as safe, collision, tailgating, or weaving.
Early DriveSafe notes on driving behavior classification, training data, model evaluation, and ways to incorporate temporal information from video.