Prerequisites |
offline | Linear algebra | | Vectors, dot products, geometric interpretations. |
offline | Vector calculus | | Taking gradients. |
offline | Probability 1 | | Discrete random variables and probability distributions, mean, variance (from Khan Academy). |
offline | Probability 2 | | Marginal and conditional distributions (from Khan Academy). |
offline | Complexity | | Basic big-Oh notation, complexity. |
offline | Optimization | | Continuous optimization, objective functions, gradient descent. |
offline | Python | | Tutorial on using Python for this course. |
General |
April 1 | Introduction | | AI history, ethics and responsibility, and what we are covering in this course. |
offline | Project | | Guidelines on how to do a project. |
Machine learning |
April 3 | Machine Learning 1 | | Linear regression and Linear classification |
April 8 | Machine Learning 2 | | SGD, feature templates, non-linear features, neural networks. |
offline | Algorithms and distribution | | Ethical frameworks related to how algorithms distribute burdens and benefits. |
April 10 | Machine Learning 3 | | Backpropagation, K-means, generalization, and best practices. |
Search |
April 15 | Search 1 | | Tree search, Dynamic programming, uniform cost search. |
April 17 | Search 2 | | Uniform cost search, A-star, A-star relaxations, (optional) Structured Perceptron. |
offline | Dual Use Technologies | | Externalities and Dual Use Technologies |
Markov Decision Processes (MDPs) |
April 22 | MDPs 1 | | Modeling, Policy evaluation, Value iteration. |
April 24 | MDPs 2 | | Reinforcement Learning, Monte Carlo, SARSA, Q-learning, exploration/exploitation, function approximation. |
offline | Aligning RL Systems | | Aligning Reinforcement Learning Systems with Human Intent. |
Games |
April 29 | Games 1 | | Adversarial games, Expectimax, Minimax, Evaluation functions, Alpha-beta pruning. |
May 1 | Games 2 | | TD-learning, Simultaneous games, Non-zero-sum games, Applications. |
offline | The AI Alignment Problem | | The AI Alignment Problem Reward Hacking and Negative Side Effects |
Constraint satisfaction problems |
May 6 | Factor Graphs | | Overview, Definitions, Examples |
May 8 | Exact and Approximate Search | | Dynamic Ordering, Arc consistency, Beam Search, Local search. |
offline | Encoding Human Values | | Encoding Human Values in AI Systems |
Markov and Bayesian Networks |
May 13 | Markov Networks and Bayesian Networks 1 | | Markov networks, Gibbs sampling, Bayesian networks definitions. |
May 15 | Bayesian Networks 2 | | Probabilistic programming, Probabilistic inference, Forward backward, Particle filtering. |
offline | AI Privacy | | Overview and Adversarial Attack Risks |
May 20 | Bayesian Networks 3 | | Supervised learning, Laplace smoothing, Expectation maximization |
Logic |
May 22 | Logic 1 | | Syntax versus semantics, Propositional logic, Inference rules. |
offline | AI Explainability and Interpretability | | Explainability and Interpretability in AI Systems |
May 29 | Logic 2 | | Modus ponens, Resolution, First order logic. |
Conclusion |
June 3 | Conclusion | | Summary of topics in CS221, future courses, and conclusion. |