Special Topics Course

Artificial intelligence holds enormous promise for accelerating scientific discovery, yet its “black box” nature and data-hungry architectures often hinder adoption in domains where interpretability, physical consistency, and sparse data prevail. This course addresses these challenges by developing a unified framework that integrates three complementary, domain-informed representations—topological features, physics-based constraints, and higher-order relational structures—into end-to-end, scalable AI models.

Networks or graphs are a natural way to represent a variety of interacting complex systems - from disease spreading and brain networks, to social networks and political networks.

In recent years, the application of Machine Learning (ML) tools has extended beyond traditional domains like image classification and Natural Language Processing (NLP) to tackle a variety of classification, detection, and control problems. More notably, the adoption of ML in addressing networking issues has emerged as a significant area of interest. However, implementing ML solutions for networked systems presents unique challenges that go beyond merely adapting tools designed for other fields.

No description currently available

This course deals with designing cryptographic systems that are conjectured to resist poly-time quantum attacks. We will discuss different (plausibly) post-quantum secure cryptographic assumptions. We will also show how to build cryptographic systems from these assumptions. We will also see cryptanalytic attacks on some cryptographic assumptions.

Pre-requisites: undergraduate course in cryptography (CMPSC 178), undergraduate course in quantum computing (CMPSC 190H).

This class is an introductory undergraduate course about networks, which can be used to study complex systems of interacting entities. The study of networks — including theory, computation, and applications — is pervasive in physics, biology, sociology, computer and information science, and myriad other fields. The study of networks is also a major part of data science.

This class is an introductory undergraduate course in machine learning. The class will briefly cover topics in regression, classification, neural networks, deep learning, convolutional and recurrent neural networks, graph neural networks, ensemble methods and reinforcement learning. The class will contain a project component.

This course will be about modern artificial intelligence and machine learning. It will start with a general overview of five fundamental neural networks (MLP, CNN, RNN, Autoencoder and Transformer) and followed by discussions of text, image and video generation techniques.

Deep neural networks have demonstrated impressive performance, yet their vulnerability to adversarial attacks has made adversarial machine learning an important topic. In this course, students will explore core principles of adversarial learning and learn how to adapt these techniques to diverse adversarial contexts. The curriculum combines lectures focused on algorithm foundations with paper presentations (by students) highlighting current state-of-the-art advances in modern AI models (e.g., large language models).

This graduate-level research course focuses on foundation models, specifically Large Language Models (LLMs). Throughout the course, we will examine the latest research publications in this rapidly evolving field, with a particular emphasis on the foundations of LLMs and their applications. Students are expected to engage in reviewing and presenting research papers, and completing a substantial course project. The primary objective of this course is to cultivate a deep understanding of LLMs and their limitations.