## CS 329M: Topics in Artificial Intelligence: Algorithms of Advanced Machine Learning

This advanced graduate course explores in depth several important classes of algorithms in modern machine learning. We will focus on understanding the mathematical properties of these algorithms in order to gain deeper insights on when and why they perform well. We will also study applications of each algorithm on interesting, real-world settings. Topics include: spectral clustering, tensor decomposition, Hamiltonian Monte Carlo, adversarial training, and variational approximation. Students will learn mathematical techniques for analyzing these algorithms and hands-on experience in using them. We will supplement the lectures with latest papers and there will be a significant research project component to the class. Prerequisites: Probability (
CS 109), linear algebra (
Math 113), machine learning (
CS 229), and some coding experience.

Last offered: Spring 2017

## CS 330: Deep Multi-task and Meta Learning

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 benleveraged 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 andnimplement 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.

Terms: Aut
| Units: 3

Instructors:
Finn, C. (PI)
;
Arumugam, D. (TA)
;
Demir, N. (TA)
;
Nair, S. (TA)
;
Rafailov, R. (TA)
;
Swofford, M. (TA)
;
Tung, A. (TA)
;
Yang, K. (TA)

## CS 345S: Data-intensive Systems for the Next 1000x

The last decade saw enormous shifts in the design of large-scale data-intensive systems due to the rise of Internet services, cloud computing, and Big Data processing. Where will we see the next 1000x increases in scale and data volume, and how should data-intensive systems accordingly evolve? This course will critically examine a range of trends, including the Internet of Things, drones, smart cities, and emerging hardware capabilities, through the lens of software systems research and design. Students will perform a comparative analysis by reading and discussing cutting-edge research while performing their own original research. Prerequisites: Strong background in software systems, especially databases (
CS 245) and distributed systems (
CS 244B), and/or machine learning (
CS 229). Undergraduates who have completed
CS 245 are strongly encouraged to attend.

Last offered: Autumn 2016

## EARTHSYS 162: Data for Sustainable Development (CS 325B, EARTHSYS 262)

The sustainable development goals (SDGs) encompass many important aspects of human and ecosystem well-being that are traditionally difficult to measure. This project-based course will focus on ways to use inexpensive, unconventional data streams to measure outcomes relevant to SDGs, including poverty, hunger, health, governance, and economic activity. Students will apply machine learning techniques to various projects outlined at the beginning of the quarter. The main learning goals are to gain experience conducting and communicating original research. Prior knowledge of machine learning techniques, such as from
CS 221,
CS 229,
CS 231N,
STATS 202, or
STATS 216 is required. Open to both undergraduate and graduate students. Enrollment limited to 24. Students must apply for the class by filling out the form at
https://goo.gl/forms/9LSZF7lPkHadix5D3. A permission code will be given to admitted students to register for the class.

Last offered: Autumn 2019
| Repeatable
for credit

## EARTHSYS 262: Data for Sustainable Development (CS 325B, EARTHSYS 162)

The sustainable development goals (SDGs) encompass many important aspects of human and ecosystem well-being that are traditionally difficult to measure. This project-based course will focus on ways to use inexpensive, unconventional data streams to measure outcomes relevant to SDGs, including poverty, hunger, health, governance, and economic activity. Students will apply machine learning techniques to various projects outlined at the beginning of the quarter. The main learning goals are to gain experience conducting and communicating original research. Prior knowledge of machine learning techniques, such as from
CS 221,
CS 229,
CS 231N,
STATS 202, or
STATS 216 is required. Open to both undergraduate and graduate students. Enrollment limited to 24. Students must apply for the class by filling out the form at
https://goo.gl/forms/9LSZF7lPkHadix5D3. A permission code will be given to admitted students to register for the class.

Last offered: Autumn 2019
| Repeatable
for credit

## EE 292D: Machine Learning on Embedded Systems

This is a project-based class where students will learn how to develop machine learning models for execution in resource constrained environments such as embedded systems. In this class students will learn about techniques to optimize machine learning models and deploy them on a device such as a Arduino, Raspberry PI, Jetson, or Edge TPUs. The class has a significant project component. Prerequisites:
CS 107(required),
CS 229 (recommended),
CS 230 (recommended).

Terms: Spr
| Units: 3

Instructors:
Asgar, Z. (PI)

## GENE 236: Deep Learning in Genomics and Biomedicine (BIODS 237, BIOMEDIN 273B, CS 273B)

Recent breakthroughs in high-throughput genomic and biomedical data are transforming biological sciences into "big data" disciplines. In parallel, progress in deep neural networks are revolutionizing fields such as image recognition, natural language processing and, more broadly, AI. This course explores the exciting intersection between these two advances. The course will start with an introduction to deep learning and overview the relevant background in genomics and high-throughput biotechnology, focusing on the available data and their relevance. It will then cover the ongoing developments in deep learning (supervised, unsupervised and generative models) with the focus on the applications of these methods to biomedical data, which are beginning to produced dramatic results. In addition to predictive modeling, the course emphasizes how to visualize and extract interpretable, biological insights from such models. Recent papers from the literature will be presented and discussed. Stude
more »

Recent breakthroughs in high-throughput genomic and biomedical data are transforming biological sciences into "big data" disciplines. In parallel, progress in deep neural networks are revolutionizing fields such as image recognition, natural language processing and, more broadly, AI. This course explores the exciting intersection between these two advances. The course will start with an introduction to deep learning and overview the relevant background in genomics and high-throughput biotechnology, focusing on the available data and their relevance. It will then cover the ongoing developments in deep learning (supervised, unsupervised and generative models) with the focus on the applications of these methods to biomedical data, which are beginning to produced dramatic results. In addition to predictive modeling, the course emphasizes how to visualize and extract interpretable, biological insights from such models. Recent papers from the literature will be presented and discussed. Students will be introduced to and work with popular deep learning software frameworks. Students will work in groups on a final class project using real world datasets. Prerequisites: College calculus, linear algebra, basic probability and statistics such as
CS 109, and basic machine learning such as
CS 229. No prior knowledge of genomics is necessary.

Terms: Aut
| Units: 3

## MATH 51: Linear Algebra, Multivariable Calculus, and Modern Applications

This course provides unified coverage of linear algebra and multivariable differential calculus, and the free course e-text connects the material to many fields. Linear algebra in large dimensions underlies the scientific, data-driven, and computational tasks of the 21st century. The linear algebra portion includes orthogonality, linear independence, matrix algebra, and eigenvalues with applications such as least squares, linear regression, and Markov chains (relevant to population dynamics, molecular chemistry, and PageRank); the singular value decomposition (essential in image compression, topic modeling, and data-intensive work in many fields) is introduced in the final chapter of the e-text. The multivariable calculus portion includes unconstrained optimization via gradients and Hessians (used for energy minimization), constrained optimization (via Lagrange multipliers, crucial in economics), gradient descent and the multivariable Chain Rule (which underlie many machine learning al
more »

This course provides unified coverage of linear algebra and multivariable differential calculus, and the free course e-text connects the material to many fields. Linear algebra in large dimensions underlies the scientific, data-driven, and computational tasks of the 21st century. The linear algebra portion includes orthogonality, linear independence, matrix algebra, and eigenvalues with applications such as least squares, linear regression, and Markov chains (relevant to population dynamics, molecular chemistry, and PageRank); the singular value decomposition (essential in image compression, topic modeling, and data-intensive work in many fields) is introduced in the final chapter of the e-text. The multivariable calculus portion includes unconstrained optimization via gradients and Hessians (used for energy minimization), constrained optimization (via Lagrange multipliers, crucial in economics), gradient descent and the multivariable Chain Rule (which underlie many machine learning algorithms, such as backpropagation), and Newton's method (an ingredient in GPS and robotics). The course emphasizes computations alongside an intuitive understanding of key ideas. The widespread use of computers makes it important for users of math to understand concepts: novel users of quantitative tools in the future will be those who understand ideas and how they fit with examples and applications. This is the only course at Stanford whose syllabus includes nearly all the math background for
CS 229, which is why
CS 229 and
CS 230 specifically recommend it (or other courses resting on it). For frequently asked questions about the differences between
Math 51 and
CME 100, see the FAQ on the placement page on the Math Department website. Prerequisite:
Math 21 or the math placement diagnostic (offered through the Math Department website) in order to register for this course.

Terms: Aut, Win, Spr, Sum
| Units: 5
| UG Reqs: GER:DB-Math, WAY-FR

Instructors:
Kim, G. (PI)
;
Lucianovic, M. (PI)
;
Spink, H. (PI)
;
Taylor, C. (PI)
;
Venkatesh, S. (PI)
;
He, J. (TA)
;
Love, J. (TA)
;
McConnell, S. (TA)
;
Nguyen, D. (TA)
;
Perlman, M. (TA)
;
Raksit, A. (TA)
;
Truong Vu, N. (TA)
;
Zavyalov, B. (TA)

## STATS 214: Machine Learning Theory (CS 229M)

How do we use mathematical thinking to design better machine learning methods? This course focuses on developing mathematical tools for answering these questions. This course will cover fundamental concepts and principled algorithms in machine learning. We have a special focus on modern large-scale non-linear models such as matrix factorization models and deep neural networks. In particular, we will cover concepts and phenomenon such as uniform convergence, double descent phenomenon, implicit regularization, and problems such as matrix completion, bandits, and online learning (and generally sequential decision making under uncertainty). Prerequisites: A solid background in linear algebra (
MATH 104,
MATH 113 or
CS 205) and probability theory (
STATS 116,
MATH 151 or
CS 109). Recommended: statistics and machine learning (
STATS 216, or
CS 229,
STATS 315A)

Terms: Win
| Units: 3

Instructors:
Ma, T. (PI)

Filter Results: