Doctor of Philosophy in Intelligent Science and Systems

Course Description

Core Courses

Literature Survey and Thesis Proposal (3 credits)

This course aims to broaden students' knowledge base and understand the development trends of scientific research in the field of intelligent sciences and systems, preparing them for selecting their doctoral dissertation topic. Students will listen to a certain number of academic reports, read a certain amount of literature, write a thesis proposal report, and obtain credits after passing the supervisor's examination.

Academic Activities (1 credit)
This course aims to help students understand the research dynamics in various fields and broaden their horizons. Students are required to participate in academic conferences, visits, lectures, and engage in research activities related to a specific topic. They are also expected to write an academic activity report. To earn credits, students need to meet a certain number of academic activity requirements and have their reports reviewed and approved by their supervisors.

Elective Courses

Intelligent Systems (3 credits)

This course covers the following main topics of intelligent systems: intelligent agent systems, heuristic search, knowledge representation, reasoning methods, bounded rationality, planning, intelligent system architectures, uncertainty handling, and machine learning (supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and lifelong learning). The course focuses on how to apply these techniques and algorithms to help humans develop intelligent systems.

Supervisory Control of Discrete Event Systems (3 credits)

This course mainly teaches the theories and methods for modeling, analysis, and control of discrete event systems. It emphasizes two mathematical formalisms: Petri nets and finite state automata. The course includes building Petri net models for resource allocation systems, as well as behavior analysis theories based on structural invariants and beacons, deadlock control methods, and performance analysis methods based on timed Petri nets. The finite state automaton method mainly teaches the supervisory control theory of discrete event systems by Ramadge-Wonham, including controllability, observability, and controller optimization.

Complex Networks (3 credits)

This course will focus on introducing basic network topology models and their properties, Internet topology characteristics and modeling, propagation mechanisms and dynamic analysis on complex networks, critical value theory of propagation on complex networks, immune strategies for complex networks, dynamics of propagation on complex networks, consecutive failures and dynamic model analysis on complex networks, including successive failure models based on coupled map lattices and search strategies in complex networks.

Engineering Optimization (3 credits)

This course mainly teaches the modeling methods and effective problem-solving methods for various engineering optimization problems. The content includes linear programming, nonlinear programming, integer programming, network programming, dynamic programming, duality theory, Lagrange methods, various intelligent optimization methods, modeling and application of optimization problems.

Stochastic Processes, Applied Statistics, and Big Data (3 credits)

This course covers topics such as random variables, conditional expectations, Markov chains, exponential distributions, Poisson processes, stationary processes, renewal theory, queuing theory, and their applications. It also includes probability and mathematical statistics, which cover concepts such as probability, random variables, joint distributions, expectations, limit theorems, sampling surveys, parameter estimation, hypothesis testing, data summarization, comparison of two samples, analysis of variance, classification data analysis, and linear least squares. Additionally, the course focuses on data mining and its various components such as data preprocessing, frequent pattern mining, classification and clustering, as well as exploring the mining of networks, complex data types, and significant application areas.

Modern Control Theory (3 credits)

This course is a fundamental required course in the field of System and Control Science, as well as an important core course in the field of automation. The course mainly teaches system analysis and design techniques based on state space methods, including expressions of system state equations, solution methods for equations, controllability and observability of systems, state space design strategies for control systems, and optimal control topics.

Algorithm Theory and Computational Complexity (3 credits)

This course is based on the fact that with the advancement of science and technology, algorithms have become essential tools for addressing a wide range of scientific and engineering challenges, serving as the foundation for intelligent systems. As such, Algorithm Theory is a discipline that delves into the design and analysis of algorithms, focusing on their computational complexity. The key contents of the course include: various fundamental approaches to algorithm design, techniques for assessing algorithmic computational complexity, problems that can be resolved using polynomial computational complexity, theories of algorithmic complexity, NP-hard issues, demonstrations of NP-hard issues, and approximation solution strategies.

Management Science (3 credits)

This course aims to introduce the techniques in management science that can be utilized in spreadsheet models to facilitate decision-making processes. The lecture materials consist of problem illustrations, problem-solving procedures, and case analyses. The case studies encompass a variety of topics, including linear programming, sensitivity analysis and simplex method, network models, integer programming, goal programming, nonlinear programming, regression analysis, discriminant analysis, time series forecasting, simulation, queueing theory, decision analysis, and project management.

Smart Material (3 credits)

This course introduces the knowledge of photocatalytic materials, including band structure engineering, design theories, and screening methods, as well as next-generation visible light-responsive photocatalysts for efficient solar energy conversion. In addition, it discusses new systems of nanoheterostructured photocatalytic materials with high quantum efficiency and novel catalytic materials, as well as applications of photocatalytic materials in carbon neutral pathways for non-fossil fuels.