MSAI 601: Probability and Statistics for AI
The course provides a foundational understanding of concepts in probability theory and statistics tailored for artificial intelligence. The course covers the basic probabilistic concepts such as probability spaces, random variables and vectors, expectation, covariance, correlation, probability distribution functions, and hypothesis testing. etc. Conditional probabilities, the Bayes formula, limit theorems, and properties of jointly distributed random variables are also covered. Students will explore practical applications of probabilistic and statistical methods within the field of artificial intelligence through hands-on exercises and real-world problems. Core.
MSAI 602: Principles of Data Science for AI
This course provides an introduction to the data science pipeline, including the processes of data collection, cleaning unstructured and messy data, data visualization, and statistical analysis. Students will also explore ethical considerations such as fairness, transparency, and bias mitigation. The course will offer students a broad overview of data science and the common tools and systems used in data science problems. Through case studies, students will consider different AI systems through the lens of data science. Core.
MSAI 603: Principles of Machine Learning for AI
This course offers an introduction to the core concepts of machine learning. Students will learn fundamental ML techniques, including supervised and unsupervised learning, neural networks, decision trees, clustering, and PCA. The course will also discuss recent applications of machine learning in AI solutions, such as computer vision, data mining, autonomous navigation, and speech recognition. Students will also gain a basic understanding of ethical AI development and AI for social good. Core.
MSAI 605: Computing Systems for AI
This course will focus on the programming, software and hardware design, and implementation issues of computing systems for machine learning and artificial intelligence applications. Students will explore a variety of topics, including basic Python program structure, functions and modules, basic I/O, object-oriented programming, database access, computer architecture, CPUs and GPUs, memory and I/O systems, virtual memory, and different processing architectures. The course will also cover AI model deployment, edge computing, and scalability challenges in large-scale AI systems. Core.
MSAI 606: Human-centered and Participatory Approaches to AI
This course will cover a broad range of issues in developing human-centered AI with a focus on participatory approaches. We will look at approaches to building AI systems that expand human capabilities, and the interplay between human and AI skills. We will explore how to make use of expertise in those communities impacted by AI systems to design them better. Topics include the fundamentals of HCI and AI, interpretability and explainability in machine learning, human-centered design for AI, adaptive user interfaces, and conversational agents. The course will teach students to design machine learning systems that are well integrated with human capabilities and concerns. Core.
MSAI 630: Safe and Trustworthy AI
Recent advances in AI have created powerful new models, but these models are not easily understood, and it is difficult to guarantee that they will behave in safe and predictable ways. In this course we will examine several key aspects of these models ranging from data privacy, secure code generation, bias and fairness, memorization and copyright infringement, poisoning and adversarial attacks on machine learning systems, reliability, robustness and safety. Core.
MSAI631: AI and Society
This course is an interdisciplinary exploration of the social impacts and ethical implications of AI. It examines the histories, social values and power dynamics shaping AI technologies, as well as how AI is reshaping culture, politics, and society. Students will develop a sociotechnical understanding of AI related to policy, education, labor, economic systems, and culture. Using approaches from the humanities and social sciences, students will develop frameworks to address ongoing challenges including digital inequality, bias, and surveillance. Students will also learn how AI has and can be used to foster positive social change. Core.
Electives
MSAI 604: Introduction to Optimization for AI
This course introduces fundamental optimization techniques essential for artificial intelligence and machine learning. Students will start with an overview of linear algebra techniques, including vector spaces, linear transformations, and eigen-decomposition, before moving to techniques in unconstrained and constrained optimization. The course will also explore global search methods, such as simulated annealing, with a focus on AI applications. Students will develop the skills to formulate and solve optimization problems, improving the efficiency and performance of AI models. Elective.
MSAI 632: Generative AI
The course will explain the fundamental principles and important techniques in building large language models (LLMs), multi-modal LLMs, and image and video generation models. The class will study Transformer architectures and their use in pretraining, and discuss methods of fine-tuning models including the use of reinforcement learning. The class will study methods of data cleaning, including efficient methods of duplicate detection. And the class will examine computing methods for large scale models that are efficient and that can run in parallel. We will also discuss image and video generation methods, such as the use of stable diffusion. Elective.
MSAI 633: AI Policy
How can regulatory strategies promote innovation while safeguarding public interest? This course provides an examination of national and international regulatory and legal frameworks governing artificial intelligence. Students will learn about topics in policy considerations, including copyright, data privacy, bias and discrimination, and the explainability and accountability of AI systems in sectors finance, healthcare, and national security. Students will also learn about contemporary developments in AI governance, including through international AI regulations, national policies, and the advocacy of standards organizations. Elective
MSAI 612: Deep Learning for AI
This course provides a comprehensive introduction to deep learning, a key driver of modern artificial intelligence, with a focus on the main features in deep neural nets and their applications in AI. Students will explore a variety of topics, including backpropagation and its importance, coding tools and their use of parallelization, autoencoders, convolutional neural networks, recurrent and recursive neural networks, and attention-based models. Students will also apply deep learning techniques to real-world problems in computer vision, natural language processing, and classification/clustering questions, gaining practical experience in building AI models. Elective.
MSAI 640: Computer Vision for AI
This course provides an in-depth introduction to computer vision, a key field in artificial intelligence that enables machines to interpret and analyze visual data. Students will explore fundamental concepts such as image filtering, correlation, object detection, image segmentation, and scene reconstruction. This course will also include discussion on facial recognition, motion tracking, and ethical considerations in vision-based AI. Students will apply computer vision techniques to real-world AI problems. Elective.
MSAI 641: Natural Language Processing for AI
This course provides students with the fundamental concepts related to computers generating and processing natural language, including morphological analysis, phrase structure, word sense disambiguation, word embedding models, and advanced deep learning architectures used in NLP. With a focus on the applications of NLP, students will explore topics related to question answering, sentiment analysis, machine translation, text summarization, and chatbot creation. Elective.
MSAI 651: Big Data Analytics for AI
This course explores the challenges, tools, and techniques for designing and implementing machine learning algorithms at scale, with a focus on AI applications. Students will learn how to configure and operate distributed computing platforms to efficiently process massive datasets. Key topics include scalable learning techniques, data streaming, data flow analytics, and machine learning on large graphs. The course covers massively parallel computing models such as MapReduce, along with methods to optimize memory, storage, and communication in parallel machine learning algorithms. Additionally, students will gain hands-on experience with SQL and NoSQL databases, distributed file systems, key-value stores, document databases, graph databases, and large-scale data visualization. Elective.