271 - 280 of 366 results for: CS

This course provides a research survey of advanced methods for robot learning in simulation, analyzing the simulation techniques and recent research results enabled by advances in physics and virtual sensing simulation. The course covers two main components: agent-environment interactions and domains for multi-agent and human-robot interaction. First, we cover agent-environment interactions by studying novel simulation environments for robotics, imitation and reinforcement learning methods, simulation for navigation and manipulation and `sim2real' techniques. In the second part, we explore models and algorithms for simulation and robot learning in multi-agent domains and human-robot interaction, studying the principles of learning for interactive tasks in which each agent collaborates to accomplish tasks. The topics include domains of social navigation, human-robot collaborative manipulation and multi-agent settings.nnThis a project-based seminar class. Projects will leverage the state-of-the-art simulation environment iGibson, in which students will develop simulations to explore learning and planning methods for diverse domains. We will provide a list of suggested projects but students might also propose an original idea. The course will cover a set of research papers with presentations by students. This is a research field in rapid transformation with exciting research lines. The goal of the class is to provide practical experience and understanding of the main research lines to enable students to conduct innovative research in this field.

Artificial intelligence is expanding the frontier of algorithm design, enabling new ways to reason about and solve complex optimization problems. This course explores how AI methods - ranging from graph neural networks and diffusion models to large language models - can be integrated into the algorithm design pipeline. We will study how to use machine learning to design new algorithms, enhance classical algorithms with data-driven components, and optimize algorithm performance in specific application domains. Topics will span both practical approaches, such as differentiable optimization and generative AI for combinatorial problems, to theoretical perspectives, including approximation guarantees and the limits of learned algorithms.

This class will provide a core overview of essential topics and new research frontiers in reinforcement learning. Planned topics include: model free and model based reinforcement learning, policy search, Monte Carlo Tree Search planning methods, off policy evaluation, exploration, imitation learning, temporal abstraction/hierarchical approaches, safety and risk sensitivity, human-in-the-loop RL, inverse reinforcement learning, learning to communicate, and insights from human learning. Students are expected to create an original research paper on a related topic. Prerequisites: CS 221 or AA 238/ CS 238 or CS 234 or CS 229 or similar experience.

AI agents need to collaborate and interact with humans in many different settings such as bots operating on social media and crowdsourcing platforms, AI assistants brokering transactions on electronic marketplaces, autonomous vehicles driving alongside humans, or robots interacting with and assisting humans in homes. Our goal in this class is to learn about and design algorithms that enable robots and AI agents to reason about their actions, interact with one another, the humans, and the environment they live in, as well as plan safe strategies that humans can trust and rely on. This is a project-based graduate course that studies algorithms in robotics, machine learning, and control theory, which can improve the state-of-the-art human-AI systems. nnRecommended: Introductory course in AI ( CS 221) and Machine Learning ( CS 229).

This interdisciplinary, project-based course at the Stanford Robotics Center (SRC) provides students with the unique opportunity to conceive, develop, and implement original creative robotics projects. This course is suitable to both undergraduate and graduate students from diverse backgrounds, including art, theater studies, music, engineering, computer science, and design. This course emphasizes creative problem-solving, collaborative teamwork, and hands-on application of robotic technologies. Students will work in groups to explore various technologies and spaces within the SRC, culminating in a public exhibition of their final projects. The course bridges artistic expression with engineering.

Language models serve as the cornerstone of modern natural language processing (NLP) applications and open up a new paradigm of having a single general purpose system address a range of downstream tasks. As the field of artificial intelligence (AI), machine learning (ML), and NLP continues to grow, possessing a deep understanding of language models becomes essential for scientists and engineers alike. This course is designed to provide students with a comprehensive understanding of language models by walking them through the entire process of developing their own. Drawing inspiration from operating systems courses that create an entire operating system from scratch, we will lead students through every aspect of language model creation, including data collection and cleansing for pre-training, transformer model construction, model training, and evaluation before deployment. Application required, apply at http://cs336.stanford.edu/

Today, anyone can train a near state-of-the-art machine learning model with a laptop, a dataset, and a few lines of code. For many applications, model building is no longer the rate-limiting step in using AI and machine learning to improve human health. In this course, we examine the other areas that need to be addressed -- from choosing the right problem, to moving technical advances into the clinical setting and making sure we are actually improving outcomes that matter. This course is geared at anyone who is looking to use AI and machine learning to make a real-world positive impact on human health. Students from all schools welcome. The course can be taken for 1 unit (lecture attendance/participation) or 2-4 units, which will include a project component. Class website: https://cs337.stanford.edu/

Within a couple of decades, or less, it is plausible that humans will create an AI that is much smarter than humans in practically all domains of human activity. We refer to such an AI as a superintelligence. The alignment problem is how to make sure that such a superintelligence acts according to its creator's intent. This course is intended for a technical audience interested in thinking about this problem. Prerequisites: one graduate-level machine learning course and one course that studies agents (e.g., AI, RL, decision analysis, economics).

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.