Stanford · 2026–27
Stanford AI / ML / Agent
course map
Stanford's AI courses are scattered across CS, EE, STATS, MS&E, CME, AA, PSYCH, and SymSys. This map turns them into one browsable path: in-depth overviews of the core courses, a full course table, and six learning paths.
Full English edition: 55 core + 70 related courses, plus paths, semester, prerequisites, and Bilibili.
All courses
Search / filter / sortArtificial Intelligence: Principles and Techniques
The overview course for AI: search, MDPs, graphical models, logic, and machine learning under one frame — making good decisions with incomplete information and limited compute.
Natural Language Processing with Deep Learning (LINGUIST 284, SYMSYS 195N)
The core NLP course from word2vec to LLMs: implement word2vec, RNNs, and attention, then build an LLM project.
Machine Learning with Graphs
Graph machine learning: GNNs, knowledge graphs, and influence maximization — turning relational data into learnable representations.
Probabilistic Graphical Models: Principles and Techniques
A systematic course on probabilistic graphical models: representation, inference, and learning — the foundation for Bayesian methods and structured prediction.
Machine Learning (STATS 229)
Stanford's flagship ML course: a single probabilistic and statistical frame that connects supervised learning, unsupervised learning, and learning theory.
Machine Learning Theory (STATS 214)
Machine learning theory: generalization bounds, non-convex optimization, and implicit regularization — why deep learning works.
Deep Learning
Ng's hands-on deep learning course: from fully connected networks to CNNs/RNNs, focused on making DL projects succeed, not just deriving formulas.
Deep Learning for Computer Vision
The best-known deep-learning course in computer vision: from kNN to Transformers, with assignments that make you hand-write backpropagation and train CNNs.
Reinforcement Learning
The core reinforcement learning course: MDPs, value iteration, policy gradients, exploration, with deep RL in the assignments.
Deep Generative Models
A theory course on deep generative models: the probabilistic foundations and training algorithms behind VAEs, GANs, flows, and diffusion.
Applied Machine Learning
A fast pass through the ML workflow, models, and evaluation for applied settings.
Deep Reinforcement Learning
RL from imitation learning to RLHF, with an emphasis on deep methods and robotics/LLM applications.
Natural Language Understanding (LINGUIST 188, LINGUIST 288, SYMSYS 195U)
Semantic representation, reasoning, and evaluation — more understanding and experiments than CS 224N.
Machine Learning for Sequence Modeling (STATS 232)
Sequence models such as HMMs, CRFs, and RNNs, plus their probabilistic foundations.
Mining Massive Data Sets
MapReduce, similarity, recommendation, graph mining, and streaming algorithms.
Deep Learning Alchemy
Training tricks, debugging, and experimental methods — explaining why things work.
Machine Learning Under Distributional Shifts
Training, evaluation, and adaptation under covariate, label, and concept shift.
Practical Machine Learning
Engineering practice for data, features, model selection, and deployment.
Deep Multi-task and Meta Learning
Multi-task learning, few-shot learning, meta-RL, and transfer learning.
Advanced Survey of Reinforcement Learning
RL theory, exploration, function approximation, and a survey of frontier directions.
Mechanistic Interpretability
Mechanistic interpretability: probing, steering, causal abstraction, and sparse autoencoders — understanding what a model is actually computing.
AI Agents and Simulations (PSYCH 186)
Simulate human society with agent-based models and generative agents, and run 'what if' experiments.
Apps With LLMs Inside
LLM application development: ship apps in sprints and understand model capability boundaries and the new programming paradigm.
Agentic AI
Turn LLMs into agents that can retrieve, reason, and execute long-horizon tasks: RAG, tools, knowledge bases, formal methods, and long-range planning.
Transformers United V6 (SYMSYS 25)
The public lecture series of the Transformer universe: one top researcher each week on LLMs, generation, robotics, and science.
Advances in Foundation Models
A panorama of foundation models: models/algorithms, systems/infrastructure, ethics/social impact, plus a hands-on project.
Self Improving AI Agents
A research seminar on self-improvement, verifiers, test-time compute, search, tools, multimodal web, and long-horizon planning.
Machine Learning from Human Preferences
The theory and practice of RLHF: from choice models and preference learning to assistance games and fairness.
Trustworthy Machine Learning: Building and evaluating agentic systems
Trustworthy agent systems: build a prototype, then use evaluation to make it reliable.
Engineering AI Agents
Engineering methodology from LLM pipelines to compound AI systems to autonomous agents: choose the problem, decompose it, pick components, and evaluate.
Language Modeling from Scratch
Build a language model from scratch: data cleaning, tokenizer, Transformer, training, scaling, evaluation, and deployment.
Artificial General Intelligence for Reasoning, Planning, and Decision Making
Reasoning, planning, and decision-making for AGI: from System 1 to System 2, multi-LLM collaboration, and temporal reasoning.
Introduction to AI Safety
From alignment and robustness to misuse risk — a basic checklist of AI safety problems.
NLP for Computational Social Science
Use NLP to measure social phenomena and causal relationships in text.
Spoken Language Processing (LINGUIST 285)
Speech recognition, synthesis, and understanding, combining signal processing and deep models.
Machine Programming
Use ML for code generation, program synthesis, and software-engineering automation.
Human Centered NLP (CS 129X)
NLP centered on users, collaboration, and interactive systems.
Aligning Superintelligence (MS&E 338)
Alignment, governance, and safety strategies for superintelligence.
Research in AI Alignment
Reading, reproduction, and project work in alignment research.
Seminar on AI Safety
Discussion of frontier papers and research directions in AI safety.
Aligning Superintelligence (CS 338)
Alignment and governance, cross-listed with CS 338.
Do Language Models Have Mental States
Discuss mind attribution to LLMs through psychological frameworks.
Statistical Models of Text and Language
Statistical modeling of text, language models, and statistical inference.
Hardware Accelerators for Machine Learning (EE 244)
ML hardware accelerators: dataflow, tensor processing, memory hierarchy, compilation, and mapping CNNs/Transformers.
Systems for Machine Learning
ML systems: efficiency across the data, training, and inference stack, with hardware-software co-design.
Machine Learning Systems Design
ML systems design: from data management to deployment and monitoring, plus privacy, fairness, and security.
AI Inference Infrastructure
AI inference infrastructure: how compound AI systems are deployed, scheduled, and scaled in the cloud.
Efficient ML Inference at Scale (EE 339)
Efficient inference at scale: model optimization, distributed serving, AI hardware, kernels, and compilers.
Fabric Architectures For AI Systems
The network fabric of AI clusters: fat trees, congestion control, and scheduling, aiming for ultra-low latency and near-100% goodput.
Securing AI Systems
Model stealing, adversarial examples, data poisoning, and the LLM system attack surface.
Machine Learning on Embedded Systems (EE 292D)
Deploy and compress models on MCUs, phones, and edge devices.
Software Techniques for Emerging Hardware Platforms (EE 349)
Write efficient software for FPGAs, accelerators, and other new hardware.
Machine Learning Systems Seminar
Discussion of frontier papers in ML systems and infrastructure.
Computer Vision: From 3D Perception to 3D Reconstruction and Beyond
Computer vision from camera models to 3D reconstruction: geometry first, deep learning second.
Generative Adversarial Networks
A dedicated GAN course: from principles to SOTA, covering training stability, evaluation, and bias.
Computer Graphics in the Era of AI
Computer graphics in the AI era: neural rendering, differentiable rendering, materials/textures, animation, and physics simulation.
Neural Models for 3D Geometry
Neural representations for 3D geometry: point clouds, meshes, implicit fields, and 3D generation.
Computational Imaging (EE 367)
Computational imaging: from convolution/denoising to inverse problems, wave optics, and end-to-end optics-algorithm co-design.
Topics in Geometric Computing - 3D and 4D Foundation Models
A special-topics course in geometric computing, recently focused on 3D/4D foundation models.
Computer Vision: Foundations and Applications
Image formation, features, geometry, and recognition — the undergraduate intro to CV.
Visual Computing Systems
Systems and hardware co-design for image and video processing.
Music and AI (MUSIC 356)
ML methods for generation, retrieval, and music understanding.
Research Seminar in Computer Vision + X
A research seminar at the intersection of vision and other fields.
Introduction to Robotics (ME 320)
Robotics fundamentals: kinematics, statics, dynamics, planning, and control.
Experimental Robotics
Experimental robotics: hands-on kinematics, dynamics, control, compliance, and collision avoidance for robot arms.
Robot Perception (EE 227)
Robot perception: vision/tactile/acoustic sensors plus algorithms, building a vision-based robot system in simulation.
Principles of Robot Autonomy I (AA 274A, EE 260A, ME 274A)
Autonomous robots I: perception, localization, SLAM, nonlinear control, motion planning, and POMDPs.
Principles of Robot Autonomy II (AA 174B, AA 274B, EE 260B, ME 274B)
Autonomous robots II: RL/optimal control, manipulation, imitation learning, human-robot interaction, and system architecture.
Sensorimotor Learning for Embodied Agents (EE 381)
Sensorimotor learning for embodied agents: from raw perception to decisions, continuously adapting to the physical world.
A Hands-On Introduction to Building AI-Enabled Robots
Build an AI robot that perceives, decides, and acts from scratch.
Algorithms for Interactive Robotics
Algorithmic problems in motion planning, grasping, and human-robot interaction.
Collaborative Robotics (ME 326)
Human-robot collaboration, shared control, and safe interaction.
Building AI-Enabled Robots
Integrate perception, planning, and control into a complete robot system.
Deep Learning in Genomics and Biomedicine (BMDS 273, GENE 236)
Deep learning in genomics and biomedicine: supervised, unsupervised, and generative models plus interpretability.
Foundation Models for Healthcare (BMDS 271, RAD 271)
Foundation models for healthcare: training, evaluating, and deploying NLP/CV/multimodal FMs.
Computational Biology: Structure and Organization of Biomolecules and Cells (BIOE 279, BIOPHYS 279, BMDS 245, CME 279)
Computational biology: protein structure prediction/design, molecular simulation, drug screening, and microscopy image analysis.
Advanced Topics in Computer Vision and Biomedicine (BMDS 276)
Advanced CV + biomedicine: visual foundation models, diffusion/generation, and vision-language deployment in healthcare.
Data Centric AI for Healthcare (BMDS 218)
Data-centric AI for healthcare: how data is generated, how bias/missingness/leakage break models, and how to fix it.
Large-Scale Neural Network Modeling for Neuroscience (PSYCH 249)
Model the brain with large-scale neural networks: vision, audition, motor control, language, and decision-making as convergent evolution.
AI and Data for Science (CS 292, PSYCH 292R, STATS 282)
AI4Science frontier lectures: one Stanford professor each week on how AI/data advances their scientific field.
Modeling Biomedical Systems (BMDS 210)
Model physiological systems with differential equations, statistics, and ML.
Algorithmic Foundations for Artificial Intelligence in Healthcare (BMDS 270)
Healthcare data, clinical prediction tasks, and the algorithms and evaluation of medical AI.
Triangulating Intelligence: Melding Neuroscience, Psychology, and AI (PSYCH 225)
Triangulate the nature of intelligence with neuroscience and psychology.
Machine Learning Methods for Neural Data Analysis (NBIO 220, STATS 220, STATS 320)
Spike sorting, behavior tracking, and neural time-series modeling.
Seminar in Artificial Intelligence in Healthcare
Frontiers of healthcare AI from hospital, industry, and investor perspectives.
AI for Science
ML topics in scientific discovery such as AlphaFold and GNoME.
Data for AI
Selection, deduplication, synthesis, attribution, and copyright of training data.
AI-Augmented Research and Development: Foundations and Practice
Bring LLMs, RAG, and agents into research workflows and validate them.
Neural Network Models of Cognition
Model cognition and development with neural networks.
A modern explainable AI approach to Theoretical Neuroscience (APPPHYS 293)
Use interpretability tools to reverse-engineer biological and artificial networks.
Machine Learning Methods for Neural Data Analysis (CS 339N, NBIO 220, STATS 320)
Statistical methods for neural data analysis, cross-listed with CS 339N.
Topics in Generative Modeling Methods in Protein Modeling and Design
Frontier papers on AlphaFold, diffusion models, and protein design.
Neuroscience and Artificial Intelligence
View AI through the history of neuroscience, then use AI tools to study the brain.
Human-Centered AI
Norvig's human-centered AI: user mental models, trustworthy design, fairness, and ethics.
Design for Artificial Intelligence (SYMSYS 195A)
Design for human-centered AI: a studio course of research, prototyping, and iteration toward responsible AI experiences.
Ethics of Artificial Intelligence
The technical side of AI ethics: fairness, interpretability, privacy, and the formal limits of these methods.
Building and Evaluating AI Products
Build an AI product and agent from 0 to 1: AI fit, latency/quality/cost, evaluation, and Demo Day.
Equity and Governance for Artificial Intelligence
Policy writing and red-team practice on algorithmic fairness, accountability, and transparency.
Algorithmic Fairness
Definitions of fairness, legal frameworks, and constrained algorithm design.
Empowering Educators via Language Technology (EDUC 473)
Use NLP to analyze teaching practice and support teachers and classrooms.
The AI Awakening: Implications for the Economy and Society
AI's impact on labor, entrepreneurship, and policy, with industry guest lectures.
Seminar on Ethical and Social Issues in Natural Language Processing (LINGUIST 287)
Bias, privacy, toxicity, manipulation, and power in NLP.
AI and Education: Foundations, Opportunities, Challenges, and Considerations
AI/edTech opportunities, challenges, and ethics for teachers.
Product Management for Engineers in the AI Age
From user problems to GTM, building a real AI product with a company partner.
The Social & Economic Impact of Artificial Intelligence (CS 22A, INTLPOL 200)
A non-technical view of AI's impact on labor, healthcare, education, and geopolitics.
Safe and Constrained AI
Safe and constrained AI: constrained optimization, safe RL, projection networks, and LLM alignment.
Efficient Generative Models
Efficient generative models: trade-offs among inference depth, compute, latency, memory, and data, with LLMs as the main case.
Statistical Learning and Data Science
An introduction to statistical learning: regression/classification, regularization, trees/ensembles, SVMs, neural networks, and clustering.
Modern Applied Statistics: Learning
Modern applied statistical learning: prediction methods, model evaluation, optimization, and generative models.
Machine Learning for Computational Engineering. (ME 343)
Kernel methods, deep learning, automatic differentiation, and physics-constrained ML.
Machine Learning for Discrete Optimization (MS&E 236)
Learn heuristics with GNNs, RL, and LLMs to improve combinatorial optimization solvers.
Decision Making under Uncertainty (AA 228)
POMDPs, dynamic programming, and RL for aerospace decision-making.
Advanced Topics in Sequential Decision Making (AA 229)
Research topics in multi-agent, partially observable, and approximate dynamic programming.
AI Measurement Science
Benchmarks, psychometrics, and validity theory for systematically evaluating AI capabilities.
Foundations of Causal Machine Learning
Semiparametric efficiency, DML, heterogeneous treatment effects, and off-policy evaluation.
AI for Algorithmic Reasoning and Optimization (MS&E 331)
Use GNNs, diffusion models, and LLMs in algorithm design and tuning.
Fundamentals of Data Science: Prediction, Inference, Causality
A unified frame for prediction, inference, and causality.
Applied Causal Inference with Machine Learning and AI (CS 288)
Causal estimation and policy evaluation with high-dimensional data.
Markov Decision Processes (EE 283)
MDPs, value iteration, policy iteration, and bandit basics.
Reinforcement Learning: Behaviors and Applications (EE 383)
Applied RL behavior design and scalable algorithms.
Foundations of Causal Machine Learning
Causal inference theory cross-listed with CS 328.
Reliability and Validity in Artificial Intelligence (STATS 357)
AI evaluation and validity, cross-listed with STATS 357.
Foundations of Reinforcement Learning with Applications in Finance (CME 241)
Model trading, hedging, and portfolio management as RL problems.
Artificial Intelligence in Financial Technology
AI applications in fintech with a team project.
Reliability and Validity in Artificial Intelligence (MS&E 330)
Evaluation, validity, distribution shift, and AI-assisted statistical inference.
No matching courses.
About this edition
All 125 course pages have an English version, translated from the Chinese notes in this repo. Official course titles, descriptions, units, terms, and instructors come from Stanford ExploreCourses. Related-course pages keep the official record and a short English one-liner.
Unofficial map. Course information is based on Stanford ExploreCourses (2026–27) and course sites; always check the official syllabus.