Machine Learning Foundations
The core concepts, methods and practice of machine learning — the basis for understanding, analyzing and applying different models.
6 units 28 notes Read 0 / 28 Sep 2026
-
L0 學習就是擬合 0 / 4 The three ingredients of supervised learning, how the loss decides what you learn, population vs. empirical risk, and gradient descent on one page.
-
L1 泛化:過擬合、欠擬合與資料切分 0 / 5 Memorizing vs. learning, the bias–variance decomposition, the roles of train / validation / test and how leakage happens, cross-validation, and reading learning curves as a diagnosis.
-
L2 實務中的最佳化 0 / 5 Minibatch gradient noise, learning rates and Adam, moving targets and EMA, regularization and early stopping, loss landscapes and initialization.
-
L3 實驗方法與超參數搜尋 0 / 4 Parameters vs. hyperparameters, grid / random / Bayesian search, nested validation and the winner's curse, seed variance and honest reporting.
-
L4 常見架構:結構對應歸納偏置 0 / 6 MLPs and universal approximation, backpropagation, convolution and translation equivariance, U-Nets and residuals, attention and Transformers, and how conditioning and time are injected.
-
L5 機率建模與生成模型的評估 0 / 4 Maximum likelihood and KL, latent variables and the ELBO, autoregressive factorization and independence assumptions, and how to evaluate a generative model.
Lit notes are what this browser has read; the record stays on this device.