课程大纲
COURSE SYLLABUS
1.
课程代码/名称
Course Code/Title
CSE5018 (CS406) /高级优化算法 Advanced Optimization Algorithms
2.
课程性质
Compulsory/Elective
Elective
3.
课程学分/学时
Course Credit/Hours
3/64
4.
授课语言
Teaching Language
英文 English
5.
授课教师
Instructor(s)
Hisao Ishibuchi
6.
是否面向本科生开放
Open to undergraduates
or not
Yes
7.
先修要求
Pre-requisites
（ 如 面 向 本 科 生 开 放 ， 请 注 明 区 分 内 容 。 If the course is open to
undergraduates, please indicate the difference.）
CSE5003 (CS419) 高级算法 Advanced Algorithms
(no difference between postgraduate and undergraduate students)
8.
教学目标
Course Objectives
This course explains a variety of advanced topics on optimization algorithms, which include hyper-heuristics,
interactive optimization, memetic algorithms, constraint handling, surrogate models, multi-tasking, transfer
optimization, noisy optimization, and multi-objective optimization. The course objective is to learn how to design
single-objective and multi-objective optimization algorithms. In addition to this main objective, students will learn how
to evaluate and compare different optimization algorithms. When we apply an optimization algorithm to a particular
application problem, we need to appropriately specify its parameters and operators. By implementing local search for
flowshop scheduling and travelling salesperson problems, students will understand that different specifications are
needed for different problems. The design of an optimization algorithm for a particular application problem can be
viewed as an optimization task. Students will learn how to handle the design of an optimization algorithm as an
optimization task in the framework of hyper-heuristics. In real-world applications, it is often the case that optimization
problems do not have a mathematically formulated objective function. Solutions are evaluated by decision makers
subjectively in some problems. In other problems, solutions are evaluated by computer simulations. In those cases,
solution evaluations are usually noisy. Some other optimization problems have multiple objectives. Students will
learn how to handle such a wide variety of optimization problems. Especially when an optimization problem has
multiple objectives, the optimization task is not to find a single optimal solution but to search for non-dominated
solutions as many as possible. Students will learn various approaches for multi-objective optimization such as
scalarizing function-based methods, constraint-based methods, reference point-based methods, and evolutionary
multi-objective optimization algorithms.
9.
教学方法
Teaching Methods
Lectures, weekly assignments and presentations.
10.
教学内容
Course Contents
Section 1
Categorization of optimization algorithms
Relation between the solution quality and the computation time
Formulation of the travelling salesman problem
Section 2
Greedy algorithm for the travelling salesman problem
General framework of local search (first move and best move)
Local search implementation for the travelling salesman problem
Section 3
Effects of the choice of a neighbourhood structure on local search behaviour
Iterated local search
Random key coding for travelling salesman problem
Section 4
Binary, permutation and random key coding for the knapsack problem
Neighbourhood structures for the knapsack problems
Various constraint handling methods
Section 5
Difficulties in performance comparison and anytime algorithms
Formulation of the flowshop scheduling problems
Heuristic algorithms for the flowshop scheduling problem
Section 6
Local search for the flowshop scheduling problem
Variable neighbourhood search
Section 7
Simulated annealing (SA) and taboo search (TS)
Data driven optimization algorithms
Section 8
Basic framework of evolutionary computation
Genetic operations (crossover and mutation)
Crossover operator for the travelling salesman problem
Section 9
Fitness evaluation and selection in evolutionary computation
Basic idea of memetic algorithms
Various implementation issues of memetic algorithms
Section 10
Hyper-heuristics (online and offline)
An implementation example of an offline hyper-heuristic algorithm
Section 11
Various topics related to optimization: Noisy optimization, experiment-based
optimization, surrogate-based optimization, transfer optimization, multi-
tasking optimization
Section 12
Basic concepts in multi-objective optimization
Comparison between traditional approach and evolutionary approach
Section 13
Evolutionary multi-objective optimization
Performance indicators for evaluating a solution set
Section 14
Visual comparison methods of solution sets
Design of evolutionary multi-objective optimization
Test problems for evolutionary multi-objective optimization
Section 15
Decomposition-based evolutionary multi-objective optimization algorithms
Indicator-based evolutionary multi-objective optimization algorithms
Section 16
Evolutionary many-objective optimization
11.
课程考核
Course Assessment
（
○
1
考核形式 Form of examination；
○
2
.分数构成 grading policy；
○
3
如面向本科生开放，请注明区分内容。
If the course is open to undergraduates, please indicate the difference.）
Class performance including attendance, assignments and presentations: 40%
Final presentation: 10%
Final examination: 50%
(no difference between postgraduate and undergraduate students)
12.
教材及其它参考资料
Textbook and Supplementary Readings
(Supplementary Readings)
Kalyanmoy Deb: Multi-Objective Optimization Using Evolutionary Algorithms, Wiley.
