Machine Learning & AI

From statistical foundations to deep learning research and reinforcement learning. All primary textbooks are free.

Prerequisites: Linear Algebra, Calculus (through multivariable), Probability & Statistics

Prerequisite Graph

graph LR
    LA[Linear Algebra] --> SL[Statistical Learning]
    PS[Prob & Stats] --> SL
    SL --> DL[Deep Learning]
    C3[Calculus 3] --> DL
    DL --> RL[Reinforcement Learning]
    PS --> RL
    DL --> FM[Foundation Models]
    FM --> LLM[LLM Systems & Inference]
    DL --> LLM
    DL --> NA[Neural Architectures]
    NA --> LLM
    DL --> TPT[Training & Post-Training]
    RL --> TPT
    TPT --> LLM
    DL --> TL[tinyllm: Build an LLM]
    LLM --> TL

Topics

#	Topic	Primary Textbook	Time
01	Statistical Learning	ISLR (free) + ESL (free)	4-5 weeks
02	Deep Learning	Goodfellow et al. (free)	6-8 weeks
03	Reinforcement Learning	Sutton & Barto (free) + Spinning Up	4-6 weeks
04	LLM Systems & Inference	ML Systems (free) + vLLM docs (free)	4-6 weeks
05	Foundation Models & Architectures	AI Engineering (Huyen) + aie-book (free)	4-5 weeks
06	Neural Architectures & DL History	CS231n (free) + AlexNet/LSTM papers (free)	3-4 weeks
07	Training & Post-Training	Chinchilla + DPO papers (free) + PyTorch/JAX docs (free)	4-6 weeks
08	tinyllm: Build an LLM from Scratch	the course: chapters you build and ol check grades	course passes 1 to 11

Quick Start

1. If you’re new to ML: start with Statistical Learning (ISLR is very accessible)
2. If you have ML basics: jump to Deep Learning (Goodfellow)
3. If you’re focused on alignment/AI safety: prioritize Deep Learning Ch 6-8, then RL + RLHF
4. For quant/trading roles: Statistical Learning + RL (especially bandits and MDPs)
5. For MLOps/infra roles: Deep Learning (transformers) → LLM Systems & Inference, pairs with Cloud Native and Data Engineering
6. For AI engineering / GenAI roles: Deep Learning → Foundation Models & Architectures (Chip Huyen) → LLM Systems & Inference
7. For foundations / interviews: pair Deep Learning with Neural Architectures & DL History — backprop, CNNs, RNNs/LSTM, and AlexNet derived from the math and coded from scratch
8. For forward deployed / inference-cloud roles: follow the Superstar FDE path (practice/bin/ol learn superstar-fde), which orders this track with the AI platform and Field Engineering tracks

See Study Plan for the 18-30 week ML/AI schedule.