MAP on C.CUI's Log

MAP on C.CUI's Loghttps://cuicaihao.github.io/tags/map/Recent content in MAP on C.CUI's LogHugoen-AUSun, 11 Oct 2026 07:00:00 +1100What Are Machine Learning Models Really Optimizing? From Gradient Descent to Large Language Modelshttps://cuicaihao.github.io/posts/2026-10-07-optimization-in-machine-learning-from-gradient-descent-to-boosting-neural-networks-and-map/Wed, 07 Oct 2026 07:00:00 +1100https://cuicaihao.github.io/posts/2026-10-07-optimization-in-machine-learning-from-gradient-descent-to-boosting-neural-networks-and-map/model.fit(X, y) hides the model, loss, constraints, and solver behind one interface. Starting with a two-dimensional quadratic, this article develops gradient descent, Newton’s method, BFGS, regularization, boosting, neural networks, MAP, and Prophet, then extends the same framework to large language model training.