Machine Learning — Study Path Agent

Chapter 1

Math & Programming Foundations

1

Linear algebra essentials for ML (vectors, matrices, norms, eigenvalues intuition)

2

Probability & statistics essentials (distributions, expectation, variance, Bayes intuition)

3

Calculus & optimization basics (gradients, chain rule, convexity intuition)

4

Python for ML (NumPy arrays, pandas DataFrames, plotting, vectorization)

Chapter 2

Data Preparation & Feature Engineering

5

Data cleaning + exploratory data analysis (EDA) workflow

6

Train/validation/test splits, leakage, and proper preprocessing pipelines

7

Feature engineering for tabular data (scaling, encoding, interactions, text/date basics)

8

Handling missingness, outliers, and class imbalance (SMOTE basics, weighting)

Chapter 3

Supervised Learning

9

Linear regression + regularization (Ridge/Lasso/Elastic Net) and diagnostics

10

Logistic regression, decision thresholds, and probability calibration

11

Support Vector Machines (margin intuition, kernels at a high level)

12

Tree-based models (Decision Trees, Random Forest, Gradient Boosting concepts)

13

Model interpretability for supervised ML (feature importance, SHAP/LIME concepts)

Chapter 4

Unsupervised & Representation Learning

14

Clustering with K-means and how to evaluate clusters (silhouette, inertia pitfalls)

15

Density & connectivity clustering (DBSCAN) and hierarchical clustering basics

16

Dimensionality reduction (PCA intuition; t-SNE/UMAP use-cases and caveats)

17

Anomaly detection basics (isolation forest intuition; reconstruction error idea)

Chapter 5

Deep Learning

18

Neural network fundamentals (forward pass, backprop, activations, loss functions)

19

Training techniques (initialization, optimizers, regularization, dropout, batch norm)

20

Convolutional Neural Networks (CNNs) for vision (filters, pooling, architectures)

21

Sequence models overview (RNN/LSTM intuition) and Transformer basics

22

Transfer learning and fine-tuning (pretrained backbones, freezing, adapters intuition)

23

Hands-on: train a small neural network end-to-end (data loaders, training loop, GPU basics)

Chapter 6

Model Evaluation, Selection & Tuning

24

Choosing metrics (MSE/MAE/R2; accuracy/precision/recall/F1/ROC-AUC/PR-AUC)

25

Cross-validation and resampling (k-fold, stratification, time-series splits basics)

26

Bias–variance tradeoff, overfitting, and learning curves (diagnosing failure modes)

27

Hyperparameter tuning (grid/random search; Bayesian optimization intuition)

28

Reproducible experiments (random seeds, baselines, ablations, result reporting)

Chapter 7

Production ML (MLOps & Deployment)

29

Serving patterns (batch vs online), latency/throughput tradeoffs, API design basics

30

Model packaging & portability (serialization pitfalls, environment management, ONNX idea)

31

Monitoring ML systems (data drift, concept drift, performance, alerting, feedback loops)

32

Pipelines + CI/CD for ML (data/versioning concepts, automated training/testing)

33

Responsible ML (fairness, privacy, security, robustness, documentation/model cards)

34

Capstone: deploy an end-to-end ML project (problem framing → model → evaluation → deploy)