Lectures on Intelligent Systems

This textbook provides the reader with an essential understanding of computational methods for intelligent systems. These are defined as systems that can solve problems autonomously, in particular problems where algorithmic solutions are inconceivable for humans or not practically executable by computers. Despite the rapidly growing applications in this field, the book avoids application details, instead focusing on computational methods that equip the reader with the methodological tools and competencies necessary to tackle current and future complex applications.

The book consists of two parts: computational intelligence methods for optimization, and machine learning. Part I begins with the concept of optimization, and introduces local search algorithms, genetic algorithms, and particle swarm optimization. Part II begins with an introduction to machine learning and covers several methods, many of which can be used as supervised learning algorithms, such as decision tree learning, artificial neural networks, genetic programming, Bayesian learning, support vector machines, and ensemble methods, plus a discussion of unsupervised learning.

This textbook is written in a self-contained style, suitable for undergraduate or graduate students in computer science and engineering, and for self-study by researchers and practitioners.

這本教科書為讀者提供了對智能系統的計算方法的基本理解。這些系統被定義為能夠自主解決問題的系統，特別是那些對人類來說算法解決方案不可想像或電腦無法實際執行的問題。儘管這個領域的應用快速增長，但本書避免了應用細節，而是專注於提供讀者所需的方法論工具和能力，以應對當前和未來的複雜應用。

本書分為兩部分：優化的計算智能方法和機器學習。第一部分從優化的概念開始，介紹了局部搜索算法、遺傳算法和粒子群優化。第二部分從機器學習的介紹開始，涵蓋了多種方法，其中許多可以用作監督學習算法，例如決策樹學習、人工神經網絡、遺傳編程、貝葉斯學習、支持向量機和集成方法，以及對無監督學習的討論。

這本教科書以自成一體的風格撰寫，適合計算機科學和工程的本科生或研究生，以及研究人員和從業人員自學使用。

Leonardo Vanneschi is a Full Professor at the Nova Information Management School (NOVA IMS) of the Universidade Nova de Lisboa, Portugal. His main research interests involve machine learning, data science, optimization, complex systems and, in particular, evolutionary computation. He has published more than 200 contributions, 11 of which have been recognized with international awards. In 2015, he received the Evo* Award for Outstanding Contribution to Evolutionary Computation in Europe. In 2020, he was included in the list of the top 2% world researchers in a study carried out by Stanford University.

Sara Silva is a Principal Investigator at the Computer Science and Engineering Research Centre (LASIGE) of the Universidade de Lisboa, Portugal. Her main research interests are machine learning and evolutionary computation, including interdisciplinary applications in the areas of remote sensing and bioinformatics. She is the author of around 100 peer-reviewed publications, having received more than 10 nominations and awards for best paper and best researcher. In 2018 she received the Evo* Award for Outstanding Contribution to Evolutionary Computation in Europe. She created the MATLAB Genetic Programming Toolbox (GPLAB).

Leonardo Vanneschi 是葡萄牙里斯本新大學Nova Information Management School (NOVA IMS)的全職教授。他的主要研究興趣包括機器學習、數據科學、優化、複雜系統，尤其是進化計算。他已發表了200多篇論文，其中11篇獲得國際獎項肯定。2015年，他獲得了Evo* Award，以表彰他對歐洲進化計算的傑出貢獻。2020年，他被斯坦福大學的一項研究列入全球前2%的研究人員名單。

Sara Silva 是葡萄牙里斯本大學計算機科學和工程研究中心(LASIGE)的首席研究員。她的主要研究興趣是機器學習和進化計算，包括遙感和生物信息學領域的跨學科應用。她是約100篇同行評審的論文的作者，並獲得了10多次最佳論文和最佳研究員的提名和獎項。2018年，她獲得了Evo* Award，以表彰她對歐洲進化計算的傑出貢獻。她創建了MATLAB遺傳編程工具箱(GPLAB)。