MDPs I: Putting Uncertainty into State Transitions | Stanford CS221 Lecture 7
Lecture 7 of Stanford CS221 Autumn 2025 follows the official material on MDPs I: Putting Uncertainty into State Transitions and makes its assumptions and limits explicit.
ai
1239 postsMDPs II: Learning Q-Values Without a Transition Model | Stanford CS221 Lecture 8
Lecture 8 of Stanford CS221 Autumn 2025 follows the official material on MDPs II: Learning Q-Values Without a Transition Model and makes its assumptions and limits explicit.
MDPs III: Differentiating Expected Return Directly | Stanford CS221 Lecture 9
Lecture 9 moves from tabular RL to function approximation, derives REINFORCE with the log-derivative identity, and connects the derivation to the executable PyTorch implementation.
Games I: From Expectimax to Minimax | Stanford CS221 Lecture 10
Lecture 10 extends single-agent search into adversarial game trees: expectimax averages chance outcomes, minimax takes the opponent's worst case, and alpha-beta removes irrelevant branches without changing the answer.
Games II: TD Learning, Simultaneous Games, and Nash Equilibria | Stanford CS221 Lecture 11
Lecture 11 first learns game values from experience with temporal-difference updates, then moves from sequential play to simultaneous games described by mixed strategies, minimax guarantees, and Nash equilibria.
Bayesian Networks I: From Joint Distributions to Factorization | Stanford CS221 Lecture 12
Lecture 12 builds a joint distribution from random variables and factors, then uses Bayesian-network factorization to express conditional independence and make conditioning and marginalization executable.
Bayesian Networks II: Gibbs Sampling and the Markov Blanket | Stanford CS221 Lecture 13
Lecture 13 replaces costly exact inference with Gibbs sampling: resample one variable at a time from a conditional determined by its Markov blanket, then approximate query probabilities with sample frequencies.
Bayesian Networks III: From Counts and Smoothing to EM | Stanford CS221 Lecture 14
Lecture 14 moves from maximum-likelihood counts and Laplace smoothing with complete data to EM, which alternates posterior responsibilities for latent variables with parameter updates.
Logic I: Models, Entailment, and SAT | Stanford CS221 Lecture 15
Lecture 15 separates propositional syntax from semantics: model checking defines entailment through satisfying assignments, SAT finds witnesses, and inference rules must be judged for both soundness and completeness.
Logic II: Quantifiers Beyond Individual Propositions | Stanford CS221 Lecture 16
Lecture 16 compresses knowledge across objects with predicates, quantifiers, and functions, then derives conclusions through substitution, unification, and definite-clause forward inference while exposing termination and completeness limits.
Language Models: From Next-Token Prediction to Generation | Stanford CS221 Lecture 17
Lecture 17 defines a language model as a chain-rule factorization of sequence probability, compares n-gram and neural conditional models, and shows how sampling, temperature, and evaluation shape generation.
AI & Society: Benefits, Misuse, Accidents, and Institutions | Stanford CS221 Lecture 18
Lecture 18 classifies AI's social effects as benefits, misuse, accidents, and structural harms, then connects fairness audits, research ethics, copyright, and platform terms to accountable institutional choices.
AI Supply Chains: Resources, Labor, and Markets Behind Models | Stanford CS221 Lecture 19
Lecture 19 uses the Economics of AI deck to connect compute, data, distribution, and organizational complements to GDP, labor, and ideas-driven growth.
Fireside Chat, Conclusion: Turning Twenty Lectures into Modeling Choices | Stanford CS221 Lecture 20
Lecture 20 is Percy Liang's fireside chat on career and research, CS221 and Stanford, and AI's future, with every attribution tied to the official video and editorial synthesis kept separate from auto-caption uncertainty.
Introduction: Why Relational Data Needs Graph Machine Learning | Stanford CS224W Lecture 1
A slide-grounded reconstruction of Fall 2025 Lecture 1, covering Course map and tools, A common language for graph data, Hand-designed features and representation learning while documenting the classroom material unavailable to self-learners.
Node Embeddings: From Random Walks to node2vec | Stanford CS224W Lecture 2
A slide-grounded reconstruction of Fall 2025 Lecture 2, covering Encoder-decoder view, Similarity and the objective, Random walks while documenting the classroom material unavailable to self-learners.
Graph Neural Networks: A First Complete Message-Passing Model | Stanford CS224W Lecture 3
A slide-grounded reconstruction of Fall 2025 Lecture 3, covering From fixed embeddings to deep encoders, The message-passing framework, Aggregation and update while documenting the classroom material unavailable to self-learners.
A General Perspective on GNNs: Turning a Model into Design Components | Stanford CS224W Lecture 4
A slide-grounded reconstruction of Fall 2025 Lecture 4, covering The GNN design space, Message, aggregation, and update, GraphSAGE while documenting the classroom material unavailable to self-learners.
GNN Augmentation and Training: Co-designing Data, Tasks, and Models | Stanford CS224W Lecture 5
A slide-grounded reconstruction of Fall 2025 Lecture 5, covering Graph-data augmentation, Feature and structural augmentation, Supervision and loss while documenting the classroom material unavailable to self-learners.
Theory of GNNs: The WL Test, GIN, and Expressive Limits | Stanford CS224W Lecture 6
A Fall 2025 slide-grounded reconstruction of Lecture 6, covering What distinguishability means, The Weisfeiler–Lehman test, An upper bound for message passing while documenting unavailable classroom material.