Skip to main content Skip to secondary navigation
Page Content

Courses

HAI believes in the integration of AI across human-centered systems and applications in a setting that can only be offered by Stanford University. Stanford’s seven leading schools on the same campus enable HAI to offer a multidisciplinary approach to education.

Learn more about AI courses at Stanford below:

Courses and Programs

    • Stanford Students
    Student teams under faculty supervision work on research and implementation of a large project in AI. State-of-the-art methods related to the problem domain. Prerequisites: AI course from 220 series, and consent of instructor.

    When

    N/A

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    This course will cover both classic findings and the latest research progress on the intersection of cognitive science, neuroscience, and artificial intelligence: How does the study of minds and machines inform and guide each other? What are the assumptions, representations, or learning mechanisms that are shared (across multiple disciplines, and what are different? How can we build a synergistic partnership between cognitive psychology, neuroscience, and artificial intelligence? We will focus on object perception and social cognition (human capacities, especially in infancy and early childhood) and the ways in which these capacities are formalized and reverse-engineered (computer vision, reinforcement learning). Through paper reading and review, discussion, and the final project, students will learn the common foundations shared behind neuroscience, cognitive science, and AI research and leverage them to develop their own research project in these areas. Recommended prerequisites: PSYCH 1, PSYCH 24/SYMSYS 1/CS 24, CS 221, CS 231N

    When

    N/A

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Advanced material is often taught for the first time as a topics course, perhaps by a faculty member visiting from another institution. May be repeated for credit.

    When

    N/A

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    The field of machine programming (MP) is concerned with the automation of software development. Given recent advances in algorithms, hardware efficiency and capacity, and an ever increasing avail- ability of code data, it is now possible to train machines to help develop software. In this course, we teach students how to build real-world MP systems. We begin by explaining the foundations of MP. Next, we analyze the current state-of-the-art MP systems (e.g., DeepMind's AlphaCode, GitHub's Copilot, Merly's MP-CodeCheck). We close with a discussion of current limitations and future utility in MP. This course also includes a six-week hands-on project, where students (as individuals or in a small group) will create their own MP system and demonstrate it to the class.

    Instructors

    • Pranay Samala
    • Justin Gottschlich

    When

    • Autumn

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    While deep learning has achieved remarkable success in supervised and reinforcement learning problems, such as image classification, speech recognition, and game playing, these models are, to a large degree, specialized for the single task they are trained for. This course will cover the setting where there are multiple tasks to be solved, and study how the structure arising from multiple tasks can be leveraged to learn more efficiently or effectively. This includes: goal-conditioned reinforcement learning techniques that leverage the structure of the provided goal space to learn many tasks significantly faster; meta-learning methods that aim to learn efficient learning algorithms that can learn new tasks quickly; curriculum and lifelong learning, where the problem requires learning a sequence of tasks, leveraging their shared structure to enable knowledge transfer. This is a graduate-level course. By the end of the course, students should be able to understand and implement the state-of-the-art multi-task learning algorithms and be ready to conduct research on these topics. Prerequisites: CS 229 or equivalent. Familiarity with deep learning, reinforcement learning, and machine learning is assumed.

    Instructors

    • Chelsea Finn
    • Suraj Nair
    • Garrett Thomas
    • Kyle Hsu
    • Yoonho Lee
    • Daniel Zeng
    • Fahim Tajwar
    • Eric Frankel
    • Max Sobol Mark
    • Eric Mitchell

    When

    • Autumn

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Understanding the human side of AI/ML based systems requires understanding both how the system-side AI works, but also how people think about, understand, and use AI tools and systems. This course will cover how what AI components and systems currently exits, along with how mental models and user models are made. These models lead to user expectations of AI systems are formed, and ultimately to design guidelines to avoid disappointing end-users by creating unintelligible AI tools that are based on a cryptic depiction of how things work. We'll also cover the ethics of AI data collection and model building, as well as how to build fair systems.

    Instructors

    • Alejandrina Gonzalez Reyes
    • Peter Norvig
    • Daniel Russell

    When

    N/A

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Artificial intelligence, specifically deep learning, stands out as one of the most transformative technologies of the past decade. AI can already outperform humans in several computer vision and natural language processing tasks. However, we still face some of the same limitations and obstacles that led to the demise of the first AI boom phase five decades ago. This research-oriented course will first review and reveal the limitations (e.g., iid assumption on training and testing data, voluminous training data requirement, and lacking interpretability) of some widely used AI algorithms, including convolutional neural networks (CNNs), transformers, reinforcement learning, and generative adversarial networks (GANs). To address these limitations, we will then explore topics including transfer learning for remedying data scarcity, knowledge-guided multimodal learning for improving data diversity, out of distribution generalization, attention mechanisms for enabling Interpretability, meta learning, and privacy-preserving training data management. The course will be taught through a combination of lecture and project sessions. Lectures on specialized AI applications (e.g., cancer/depression diagnosis and treatment, AI/VR for surgery, and health education) will feature guest speakers from academia and industry. Students will be assigned to work on an extensive project that is relevant to their fields of study (e.g., CS, Medicine, and Data Science). Projects may involve conducting literature surveys, formulating ideas, and implementing these ideas. Example project topics are but not limited to 1) knowledge guided GANs for improving training data diversity, 2) disease diagnosis via multimodal symptom checking, and 3) fake and biased news/information detection.

    When

    • Spring

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Artificial Intelligence (AI) has the potential to drive us towards a better future for all of humanity, but it also comes with significant risks and challenges. At its best, AI can help humans mitigate climate change, diagnose and treat diseases more effectively, enhance learning, and improve access to capital throughout the world. But it also has the potential to exacerbate human biases, destroy trust in information flow, displace entire industries, and amplify inequality throughout the world. We have arrived at a pivotal moment in the development of the technology in which we must establish a foundation for how we will design AI to capture the positive potential and mitigate the negative risks. To do this, building AI must be an inclusive, interactive, and introspective process guided by an affirmative vision of a beneficial AI-future. The goal of this interdisciplinary class is to bridge the gap between technological and societal objectives: How do we design AI to promote human well-being? The ultimate aim is to provide tools and frameworks to build a more harmonious human society based on cooperation toward a shared vision. Thus, students are trained in basic science to understand what brings about the conditions for human flourishing and will create meaningful AI technologies that aligns with the PACE framework: 1) has a clear and meaningful purpose, 2) augments human dignity and autonomy, 3) creates a feeling of inclusivity and collaboration, 4) creates shared prosperity and a sense of forward movement (excellence). Toward this end, students work in interdisciplinary teams on a final project and propose a solution that tackles a significant societal challenge by leveraging technology and frameworks on human thriving.

    When

    N/A

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    In this seminar, we will focus on the challenges in the design of safe and verified AI-based systems. We will explore some of the major problems in this area from the viewpoint of industry and academia. We plan to have a weekly seminar speaker to discuss issues such as verification of AI systems, reward misalignment and hacking, secure and attack-resilient AI systems, diagnosis and repair, issues regarding policy and ethics, as well as the implications of AI safety in automotive industry. Prerequisites: There are no official prerequisites but an introductory course in artificial intelligence is recommended.

    When

    N/A

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Artificial intelligence is poised to make radical changes in healthcare, transforming areas such as diagnosis, genomics, surgical robotics, and drug discovery. In the coming years, artificial intelligence has the potential to lower healthcare costs, identify more effective treatments, and facilitate prevention and early detection of diseases. This class is a seminar series featuring prominent researchers, physicians, entrepreneurs, and venture capitalists, all sharing their thoughts on the future of healthcare. We highly encourage students of all backgrounds to enroll (no AI/healthcare background necessary). Speakers and more at shift.stanford.edu/healthai.

    When

    • Autumn

    Subject

    • CS

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    A historically-informed and multidisciplinary approach to designing emerging technologies like AI, the metaverse, NFTs, IOT, and the systems in which they exist to be equitable.Throughout history, innovations in science and technology, while bold and visionary, have often resulted in catastrophic consequences for Indigenuous, Black, immigrant and other historically oppressed communities, and the environment. Today's emerging technologies, which span everything from deep fakes to self-driving vehicles, have incredible capabilities, and at the same time are plagued with algorithmic bias and lack accountability. What can we learn from our precarious past that we are not learning today, but need to? This class welcomes the curious and the creative, from interdisciplinary fields including design, computer science, art, history, political science, ethics, feminist and gender studies and professionals of diverse backgrounds. Through a variety of hands-on design projects including historical excavations, speculative fiction and world building, students will learn how to create prototypes of emerging technologies and evaluate their implications on diverse communities, the environment, and our past and future selves.

    Instructors

    • Ariam Mogos

    When

    N/A

    Subject

    • DESIGN

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    How do we design artificial systems that learn as we do early in life -- as "scientists in the crib" who explore and experiment with our surroundings? How do we make AI "curious" so that it explores without explicit external feedback? Topics draw from cognitive science (intuitive physics and psychology, developmental differences), computational theory (active learning, optimal experiment design), and AI practice (self-supervised learning, deep reinforcement learning). Students present readings and complete both an introductory computational project (e.g. train a neural network on a self-supervised task) and a deeper-dive project in either cognitive science (e.g. design a novel human subject experiment) or AI (e.g. implement and test a curiosity variant in an RL environment). Prerequisites: python familiarity and practical data science (e.g. sklearn or R).

    Instructors

    • Nick Haber

    When

    • Spring

    Subject

    • EDUC

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    The seminar will feature guest lectures from industry and academia to discuss the state of the affairs in the field of Robotics, Artificial Intelligence (AI), and how that will impact the future Education. The time of robotics/AI are upon us. Within the next 10 to 20 years, many jobs will be replaced by robots/AI. We will cover hot topics in Robotics, AI, how we prepare students for the rise of Robotics/AI, how we Re-design and Re-invent our education to adapt to the new era

    Instructors

    • Li Jiang

    When

    • Winter

    Subject

    • EDUC

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Applied linear algebra and linear dynamical systems with applications to circuits, signal processing, communications, and control systems. Topics: least-squares approximations of over-determined equations, and least-norm solutions of underdetermined equations. Symmetric matrices, matrix norm, and singular-value decomposition. Eigenvalues, left and right eigenvectors, with dynamical interpretation. Matrix exponential, stability, and asymptotic behavior. Multi-input/multi-output systems, impulse and step matrices; convolution and transfer-matrix descriptions. Control, reachability, and state transfer; observability and least-squares state estimation. Prerequisites: Linear algebra and matrices as in ENGR 108 or MATH 104; ordinary differential equations and Laplace transforms as in EE 102B or CME 102.

    Instructors

    • Nick Landolfi
    • Nikhil Devanathan
    • Nitish Gudapati
    • Amirhossein Afsharrad
    • Sanjay Lall

    When

    • Summer
    • Autumn

    Subject

    • EE

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Lecture/studio course exploring arts and humanities scholarship and practice engaging with, and generated by, emerging emerging and exponential technologies. Our course will explore intersections of art and artificial intelligence with an emphasis on social impact and racial justice. Open to all undergraduates.

    Instructors

    • Michele Elam
    • Camille Utterback

    When

    • Spring

    Subject

    • ENGLISH

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    This course explores cutting-edge disputes in the ethics, law and politics of artificial intelligence. We will examine the relation between foundational questions about fairness, autonomy, corporate responsibility, and the value of human life; and practical questions about the ethical design and regulation of emerging technologies. Topics include superintelligence and existential risk, explainable intelligent systems, nudging and targeted advertising, and algorithmic fairness.

    When

    • Spring

    Subject

    • ETHICSOC

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    What is AI and why is it sometimes biased ? How will AI affect medicine to help us but also what are the conditions in which it may harm us. 95% of single-gene diseases we know of have no effective treatment yet if we change a defective one how might that affect a species in the long term ? Is DNA 'the code of life?' Or is the 'code of life' the whole living organism in its complex, dynamic relationship with its environment? Will Earth one day be populated by beings who are different from us in their cognitive and physical abilities. This course will look at the intersection of AI and Genetics to analyze advances that could be made but also ethical questions that should be asked. The course is designed to be accessible to many disciplines and there are no pre-requisites.

    Instructors

    • Michael Snyder
    • Artem Trotsyuk
    • Ronjon Nag

    When

    N/A

    Subject

    • GENE

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Topics: This is a graduate level introduction to bioinformatics and computational biology, algorithms for alignment of biological sequences and structures, computing with strings, phylogenetic tree construction, hidden Markov models, basic structural computations on proteins, protein structure prediction, molecular dynamics and energy minimization, statistical analysis of 3D biological data, integration of data sources, knowledge representation and controlled terminologies for molecular biology, microarray analysis, chemoinformatics, pharmacogenetics, network biology. Note: For Fall 2021, Dr. Altman will be away on sabbatical and so class will be taught from lecture videos recorded in fall of 2018. The class will be entirely online, with no scheduled meeting times. Lectures will be released in batches to encourage pacing. A team of TAs will manage all class logistics and grading. Firm prerequisite: CS 106B.

    Instructors

    • Russ Altman
    • Daisy Ding
    • Samson Mataraso
    • Josiah Aklilu
    • Kristy Carpenter

    When

    • Autumn

    Subject

    • GENE

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Artificial Intelligence (AI) has the potential to drive us towards a better future for all of humanity, but it also comes with significant risks and challenges. At its best, AI can help humans mitigate climate change, diagnose and treat diseases more effectively, enhance learning, and improve access to capital throughout the world. But it also has the potential to exacerbate human biases, destroy trust in information flow, displace entire industries, and amplify inequality throughout the world. We have arrived at a pivotal moment in the development of the technology in which we must establish a foundation for how we will design AI to capture the positive potential and mitigate the negative risks. To do this, we must be intentional about human-centered design because, ¿Only once we have thought hard about what sort of future we want, will we be able to begin steering a course toward a desirable future. If we don¿t know what we want, we¿re unlikely to get it.¿ Thus, building AI must be an inclusive, interactive, and introspective process guided by an affirmative vision of a beneficial AI-future. The goal of this interdisciplinary class is to bridge the gap between technological and societal objectives: How do we design AI to promote human well-being? The ultimate aim is to provide tools and frameworks to build a more harmonious human society based on cooperation toward a shared vision. Thus, students are trained in basic science to understand what brings about the conditions for human flourishing and will create meaningful AI technologies that aligns with the PACE framework:·has a clear and meaningful purpose ·augments human dignity and autonomy ·creates a feeling of inclusivity and collaboration·creates shared prosperity and a sense of forward movement (excellence)Toward this end, students work in interdisciplinary teams on a final project and propose a solution that tackles a significant societal challenge by leveraging technology and frameworks on human thriving.

    When

    N/A

    Subject

    • GSBGEN

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Recent advances in computing may place us at the threshold of a unique turning point in human history. Soon we are likely to entrust management of our environment, economy, security, infrastructure, food production, healthcare, and to a large degree even our personal activities, to artificially intelligent computer systems. The prospect of "turning over the keys" to increasingly autonomous systems raises many complex and troubling questions. How will society respond as versatile robots and machine-learning systems displace an ever-expanding spectrum of blue- and white-collar workers? Will the benefits of this technological revolution be broadly distributed or accrue to a lucky few? How can we ensure that these systems are free of algorithmic bias and respect human ethical principles? What role will they play in our system of justice and the practice of law? How will they be used or abused in democratic societies and autocratic regimes? Will they alter the geopolitical balance of power, and change the nature of warfare? The goal of CS22a is to equip students with the intellectual tools, ethical foundation, and psychological framework to successfully navigate the coming age of intelligent machines.

    Instructors

    • Jerry Kaplan

    When

    N/A

    Subject

    • INTLPOL

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    Explores the relation between technology, war, and national security policy with reference to current events. Course focuses on current U.S. national security challenges and the role that technology plays in shaping our understanding and response to these challenges, including the recent Russia-Ukraine conflict. Topics include: interplay between technology and modes of warfare; dominant and emerging technologies such as nuclear weapons, cyber, sensors, stealth, and biological; security challenges to the U.S.; and the U.S. response and adaptation to new technologies of military significance.

    Instructors

    • Herb Lin
    • Rhea Kumar
    • Niv Rajesh
    • Theo Velaise
    • Antone Cruz
    • Vaidehi Bhaskara
    • Alexis Legrand

    When

    • Autumn

    Subject

    • INTLPOL

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    This course explores how new technologies pose challenges and create opportunities for the United States to compete more effectively with rivals in the international system with a focus on strategic competition with the People's Republic of China. In this experiential policy class, you will address a priority national security challenge employing the "Lean" problem solving methodology to validate the problem and propose a detailed technology informed solution tested against actual experts and stakeholders in the technology and national security ecosystem. The course builds on concepts presented in MS&E 193/293: Technology and National Security and provides a strong foundation for MS&E 297: Hacking for Defense.

    Instructors

    • Joe Felter
    • Raj Shah
    • Steve Blank

    When

    • Autumn

    Subject

    • INTLPOL

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    (Cross-listed with LAW 4052.) Course surveys current and emerging legal and governance problems related to humanity's relationship to artificially constructed intelligence. To deepen students' understanding of legal and governance problems in this area, course explores definitions and foundational concepts associated with AI, likely pathways of AI's evolution, different types of law and policy concerns raised by existing and future versions of AI, and the distinctive domestic and international political economies of AI governance. Course also covers topics associated with the design and development of AI as they relate to law and governance, such as measuring algorithmic bias and explainability of AI models. Cross-cutting themes include: how law and policy affect the way important societal decisions are justified; the balance of power and responsibility between humans and machines in different settings; the incorporation of multiple values into AI decision-making frameworks; the interplay of norms and formal law; technical complexities that may arise as society scales deployment of AI systems; AI's implications for transnational law and governance and geopolitics; and similarities and differences to other domains of human activity raising regulatory trade-offs and affected by technological change. Note: Course is designed both for students who want a survey of the field and lack any technical knowledge, as well as students who want to gain tools and ideas to deepen their existing interest or technical background in the topic. Taught by a sitting judge, a former EU Parliament member, and a law professor, and conceived to serve students with interest in law, business, public policy, design, and ethics. Course includes lectures, practical exercises, and student-led discussion and presentations. CONSENT APPLICATION: To accommodate as many students as possible, please fill out the following application by March 12, 2021 in order to facilitate planning and confirm your level of interest: https://docs.google.com/forms/d/e/1FAIpQLSfwRxaM1omTsJmK9k0gksdS5jBPRz-YCuYhRUpDlVXXglDHjg/viewform. Applications received after deadline will be considered on a rolling basis pending space. Application also available on SLS website (Click Courses at the bottom of homepage and then click Consent of Instructor Forms).

    When

    N/A

    Subject

    • INTLPOL

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    We will consider the developing legal and ethical problems of robots and artificial intelligence (AI), particularly self-directed and learning AIs. How do self-driving cars (or autonomous weapons systems) value human lives? How do we trade off accuracy against other values in predictive algorithms? At what point should we consider AIs autonomous entities with their own rights and responsibilities? And how can courts and legislatures set legal rules robots can understand and obey? This discussion seminar will meet four times during the Fall quarter. Meeting dates and times to be arranged by instructor. Elements used in grading: Attendance and class participation.

    When

    N/A

    Subject

    • LAW

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A

    • Stanford Students
    People often tend to think of technology as value neutral, as essentially objective tools that can be used for good or evil, particularly when questions of race and racial justice are involved. But the technologies we develop and deploy are frequently shaped by historical prejudices, biases, and inequalities and thus may be no less biased and racist than the underlying society in which they exist. In this discussion group, we will consider whether and how racial and other biases are present in a wide range of technologies, such as "risk assessment" algorithms for bail, predictive policing, and other decisions in the criminal justice system; facial recognition systems; surveillance tools; algorithms for medical diagnosis and treatment decisions; online housing ads that result in "digital redlining;" programs that determine entitlement to credit or public benefits and/or purport to detect fraud by recipients; algorithms used in recruiting and hiring; digital divide access gaps; and more. Building on these various case studies, we will seek to articulate a framework for recognizing both explicit and subtle anti-black and other biases in tech and understanding them in the broader context of racism and inequality in our society. Finally, we will discuss how these problems might be addressed, including by regulators, legislators, and courts as well as by significant changes in mindset and practical engagement by technology developers and educators. Elements used in grading: Full attendance, reading of assigned materials, and active participation. Class meets 4:30 PM-6:00 PM on Sept. 29, Oct. 13, Oct. 27, Nov. 10.

    Instructors

    • Phillip Malone

    When

    • Autumn

    Subject

    • LAW

    Delivery Method

    N/A

    Time Commitment

    Academic Quarter

    Earned Outcome

    N/A

    Cost

    N/A