AA228 / CS238: Decision Making under Uncertainty
Autumn Quarter 2014-pres
This course introduces decision making under uncertainty from a computational perspective, and provides an overview of the necessary tools for building autonomous and decision-support systems. Following an introduction to probabilistic models and decision theory, the course will cover computational methods for solving decision problems with stochastic dynamics, model uncertainty, and imperfect state information. Topics include: Bayesian networks, influence diagrams, dynamic programming, reinforcement learning, and partially observable Markov decision processes. Applications cover: air traffic control, aviation surveillance systems, autonomous vehicles, and robotic planetary exploration. Prerequisites: basic probability and fluency in a high-level programming language.
AA228V / CS238V: Validation of Safety Critical Systems
Winter Quarter 2025 (lectures by Sydney Katz)
Before deploying autonomous decision-making systems in high-stakes applications, it is important to ensure that they will operate as intended. This course presents a variety of mathematical concepts and algorithms that can be used to validate their performance in simulation. The course first introduces a framework for setting up validation problems using topics from model fitting, model validation, and property specification. The course then covers sampling-based validation techniques for tasks such as falsification and probability of failure estimation. The course concludes with an overview of formal methods for tasks such as reachability analysis. Topics include but are not limited to: mathematical modeling, temporal logic specifications, optimization-based falsification, Markov chain Monte Carlo, importance sampling, reachability analysis, model checking, satisfiability, and explainability. Applications are drawn from air traffic control, autonomous systems, and self-driving cars.
AA229 / CS239: Advanced Topics in Sequential Decision Making
Winter Quarter 2015, 2016, 2018, 2020, 2022, 2024
This course surveys recent research advances in intelligent decision making for dynamic environments from a computational perspective. It will explore efficient algorithms for single and multiagent planning in situations where a model of the environment may or may not be known. Topics include partially observable Markov decision processes, approximate dynamic programming, and reinforcement learning. This course discusses new approaches for overcoming challenges in generalization from experience, exploration of the environment, and model representation so that these methods can scale to real problems in a variety of domains including aerospace, air traffic control, and robotics. Students are expected to produce an original research paper on a relevant topic. Prerequisites: AA228/CS238 or CS221.
AA222: Introduction to Multidisciplinary Design Optimization
Spring Quarter 2014-pres
Design of aerospace systems within a formal optimization environment. Mathematical formulation of the multidisciplinary design problem (parameterization of design space, choice of objective functions, constraint definition); survey of algorithms for unconstrained and constrained optimization and optimality conditions; description of sensitivity analysis techniques. Hierarchical techniques for decomposition of the multidisciplinary design problem; use of approximation theory. Applications to design problems in aircraft and launch vehicle design. Prerequisites: multivariable calculus; familiarity with a high-level programming language: FORTRAN, C, C++, MATLAB, Python, or Julia.
AA120Q: Building Trust in Autonomy
Winter Quarter 2017, 2019, 2021, 2025
Major advances in both hardware and software have accelerated the development of autonomous systems that have the potential to bring significant benefits to society. Google, Tesla, and a host of other companies are building autonomous vehicles that can improve safety and provide flexible mobility options for those who cannot drive themselves. On the aviation side, the past few years have seen the proliferation of unmanned aircraft that have the potential to deliver medicine and monitor agricultural crops autonomously. In the financial domain, a significant portion of stock trades are performed using automated trading algorithms at a frequency not possible by human traders. How do we build these systems that drive our cars, fly our planes, and invest our money? How do we develop trust in these systems? What is the societal impact on increased levels of autonomy?
Executive Education: Harnessing AI for Breakthrough Innovation and Strategic Impact
Weeklong programs in October, April, and July
Make more informed and strategic decisions for your organization with a deeper understanding of the artificial intelligence landscape and its real-world applications. Artificial Intelligence is transforming almost every dimension of business, government, and society. Are you up to speed? Do you understand the risks and rewards? Is there a strategic case for making it part of your business? Harnessing AI for Breakthrough Innovation and Strategic Impact will give you the technology lay of the land — from generative AI and full autonomy to whatever comes next. You’ll discover best practices and lessons learned from industry leaders who are actually using AI. And, you’ll explore opportunities and consequences to help you decide if — and where — investing in AI makes sense for your organization. Faculty from Stanford Graduate School of Business, Stanford Institute for Human-Centered Artificial Intelligence (HAI), School of Engineering, Law School, Medical School, and School of Humanities and Sciences will join forces to demystify AI in a non-technical way, and provide strategies and frameworks to help your organization innovate and take the lead. It’s a powerful combination of technological expertise and business innovation you’ll find only at Stanford, in the heart of Silicon Valley.