Postgraduate Courses

This course helps students to get basic knowledge about deep neural networks, helping them to understand basic concepts, capabilities and challenges of deep neural networks.

This course covers recent developments in deep learning. Topics include meta learning, model compression, federated learning, representation learning, explainable AI, adversarial attack and defense, and advances in deep learning theory.

This course covers recent developments in deep reinforcement learning. Topics include reinforcement learning basics, deep Q-learning, policy gradients, actor-critic algorithms, model-based reinforcement learning, imitation learning, inverse reinforcement learning, hierarchical reinforcement learning, and multi-agent reinforcement learning.

This course covers popular topics in computer vision, which includes high-level tasks like image classification, object detection, image segmentation, and low-level tasks like image generation, image enhancement, image-to-image translation, etc.

This is a task-oriented yet interaction-based course, which aims to scrutinize the recent trends and challenges in visual intelligence tasks (from the image restoration to 3D vision tasks). This course will follow the way of flipped-classroom manner where the lecturer teaches the basics; meanwhile, the students will also be focused on active discussions, presentations (lecturing), and hands-on research projects under the guidance of the lecturer in the whole semester. Through this course, students will be equipped with the capability to critically challenge the existing methodologies/techniques and hopefully make breakthroughs in some new research directions.

This course covers recent developments in machine learning on graph-structured data. Topics include network embedding, graph neural networks, knowledge graph embedding, generative models for graphs, scalable graph neural networks, explainable graph neural networks, and their applications.

This course is a hands-on introduction to the algorithms and programming for visual learning problems of the intelligent mobile systems, such as self-driving cars, in the era of industry 4.0. This course is highlighted by practical programming assignments and projects. This course will first introduce the basic principles of machine/deep learning and computer vision. Then, the specific visual learning tasks crucial for achieving intelligent mobile systems (e.g., object detection, semantic segmentation, optical flow, stereo/depth estimation, pose estimation, lane detection, traffic sign detection) will be covered. Finally, the latest development of novel sensor-based vision and AI-based augmented reality (AR) /metaverse technologies for intelligent mobile systems will also be introduced.

This course will introduce the fundamental principles, uses, and technical details of data mining techniques by lectures and real-world case studies. The emphasis is on understanding the basic data mining techniques and their applications. We will discuss the mechanics of how data analytics techniques work as is necessary to understand the fundamental concepts and real-world applications.

This course covers how to program with Python and use it to solve practical problems in Artificial Intelligence. Topics include basic Python usage (e.g., syntax, data structure, etc.) and important packages for data analysis and machine learning applications (e.g., NumPy, SciPy, etc.). The students will be guided to practice on simple artificial intelligence tasks.

This course aims to provide students with an overview of Artificial Intelligence (AI) principles and techniques. Key topics include machine learning, search, game theories, Markov decision process, constraint satisfaction problems, Bayesian networks, etc. Through this course, students will learn and practice the foundational principles, techniques and tools to tackle new AI problems.

This course introduces potential security and privacy vulnerabilities in Artificial Intelligence (AI) and covers basic and advanced protections. Topics include security and privacy risks in AI technologies, the goal of C.I.A. (Confidentiality, Integrity and Availability) in AI technologies, basic and advanced cryptography, protocol designs for AI security and privacy, etc.

Learning to make good decisions is one of the keys to autonomous systems. This course will focus on Reinforcement Learning (RL), a currently very active subfield of artificial intelligence, and it will discuss selectively a number of algorithmic topics including Markov Decision Process, Q-Learning, function approximation, exploration and exploitation, policy search, imitation learning, model-based RL and optimal control. This course provides both the foundations and techniques for developing RL and deep RL algorithms that interact with physical environments, and real application cases of RL will be introduced. Basic knowledge of machine learning and mathematical optimization are expected for this course.

This course aims to provide students with key principles and algorithms to build modern autonomous AI systems. Key topics include machine perception, planning and decision-making algorithms. Through this course, students will learn and practice the foundational principles, techniques, and tools to build new autonomous AI systems.

This course covers typical algorithms and data structures. Topics include core methodologies of algorithm design, standard data structures, and typical algorithms and data structures that have been widely adopted for solving different problems, covering from fundamental ones (e.g., searching and sorting algorithms) to more advanced ones (e.g., graph algorithms, number theory algorithms, FFT).

This course will explore the basic knowledge and the latest advances of Deep Learning based methods in the medical field, with special attention to challenges and opportunities for Medical AI. This course will provide students the opportunity to learn skills to train/learn/develop Deep Learning models from medical data with several case studies. It covers knowledge on Digital Medical Imaging, supervised learning, semi-supervised learning, and unsupervised learning related with Medical AI research. Importantly, some hot topics on "AI for Medical" will be introduced.

Machine learning is a cornerstone of AI. The course targets beginners who will learn basic and rigorous machine learning methods, including linear regression, logistic regression, decision trees, naïve Bayes, SVM, unsupervised learning, neural networks, graphical models, EM algorithm. The students will be able to enter more advanced machine learning courses after taking this course.

Artificial Intelligence technologies have been maturing and are deployed in real-world applications, such as healthcare, entertainment, business, scientific research, military, etc. In all these domains, the decisions made by AI algorithms can critically impact individuals, organizations and society. The designers, auditors, and users of AI technologies thus need to be equipped with the capabilities to understand, analyze, and eventually discipline these algorithms in the broader contexts. This course will introduce students to the latest research of responsible AI and explore these capabilities in both theoretical and practical ways. Topics include but are not limited to theories and algorithms of secure machine learning, fair machine learning, interpretable AI, and case studies involving natural language processing, computer vision, and reinforcement learning.

This course focuses on the Artificial Intelligence (AI) techniques and applications in multimodal tasks, which involve processing, fusing, and generating contents from multiple data modalities, such as images, videos, text etc. The course will cover the challenges, state-of-the-art methods, as well as hands-on experience in implementing and evaluating multi-modal deep learning models.

In the era of large-scale deep learning models, multimodal learning based on speech, text, and images is gaining increasing prominence. It holds the potential to facilitate cross-domain applications, improve human-computer interaction, and advance innovation in the field of AI. This course will provide an in-depth exploration of applied deep learning techniques, focusing on their applications in speech processing, natural language understanding, and multimodal data analysis. Students will gain practical experience in building deep learning models for various tasks, including speech recognition, language translation, image analysis, and more. The course covers fundamental concepts, algorithms, and tools in the field of deep learning and emphasizes hands-on projects and real-world applications.

This course provides a comprehensive introduction to the field of quantum computing. Students will explore fundamental concepts such as quantum bits, quantum circuits, and quantum algorithms. Advanced topics including quantum error correction, quantum information processing, and quantum machine learning will also be covered.

Selected topics in Artificial Intelligence (AI) of current interest of current interest in emerging areas and not covered by existing courses. May be repeated for credit if different topics are covered. May be graded by letter or P/F for different offerings.

Covers emerging topics of machine learning. Potential topics include machine learning and cognitive science, transfer learning, multi-task learning, active learning, lifelong learning, assemble learning, and advances in deep learning. Graded P or F.

The course will provide students with the opportunity to gain relevant knowledge, skills, and experience while establishing important connections in the field. Graded P or F.

An independent research project carried out under the supervision of a faculty member. Graded P or F.

Series of seminars presenting research problems currently under investigation, presented by faculty, students, and visiting speakers. Students are expected to attend regularly. Graded P or F.

Series of seminars presenting research problems currently under investigation, presented by faculty, students, and visiting speakers. Students are expected to attend regularly. Continuation of AIAA 6101. Graded P or F.

Master's thesis research supervised by co-advisors from different disciplines. A successful defense of the thesis leads to the grade Pass. No course credit is assigned.

Original and independent doctoral thesis research supervised by co-advisors from different disciplines. A successful defense of the thesis leads to the grade Pass. No course credit is assigned.