Course positioning
CS 221 is the entry point to Stanford's AI track, and the best way to see that AI is more than model tuning. It uses one mathematical language (probability + algorithms) to describe how agents make decisions with incomplete information and limited computation, spanning classical symbolic AI through modern machine learning.
What it covers
- Search and constraint satisfaction: DFS/BFS/A*, heuristics, CSPs, local search
- Games and adversarial search: minimax, alpha-beta, expected utility
- MDPs and reinforcement learning: value iteration, policy iteration, Q-learning basics
- Probabilistic graphical models: Bayesian networks, HMMs, variable elimination, sampling
- Machine learning: naive Bayes, perceptron, logistic regression, losses and regularization
- Logic and knowledge representation: propositional logic, first-order logic, inference
- Capstone project: combine the modules into a working AI system
Prerequisites
CS 103 (or 103B/X), CS 106B (or 106X), CS 109, and CS 161. The course explicitly says it will not reteach these foundations.
Assessment & workload
Typically about five written/programming assignments plus a midterm, a final, and a capstone project; the workload is on the heavy side for a Stanford course. Check the current syllabus for exact weights.
Who it's for
Undergrads or students switching majors who want a complete AI map; anyone planning to take CS 229/224N/234 next; engineers who want to fill the classical-AI gap.
Study advice
Brush up on probability (CS 109) and algorithms (CS 161) first, or the assignments will hurt. Write the search and MDP parts from scratch rather than only calling libraries.
Related courses
Official course description
Artificial intelligence (AI) has had a huge impact in many areas, including medical diagnosis, speech recognition, robotics, web search, advertising, and scheduling. This course focuses on the foundational concepts that drive these applications. In short, AI is the mathematics of making good decisions given incomplete information (hence the need for probability) and limited computation (hence the need for algorithms). Specific topics include search, constraint satisfaction, game playing,n Markov decision processes, graphical models, machine learning, and logic. Prerequisites: CS 103 or CS 103B/X, CS 106B or CS 106X, CS 109, and CS 161 (algorithms, probability, and object-oriented programming in Python). We highly recommend comfort with these concepts before taking the course, as we will be building on them with little review.