[1/2] Grokking Artificial Intelligence Algorithms (Growing algorithms of artificial intelligence) (AI column)

I finished reading this book about artificial intelligence algorithms and can say that in the almost five years since the publication of the book, something has changed:) The author focused on the story of fundamental algorithms: search, evolutionary algorithms, swarm intelligence, classical methods of machine learning, and at the end briefly talked about ANN. (artificial neural networks) I moved on to reinforcement learning. But since then, there has been a significant breakthrough in the development of LLM. (large-language models) Gen AI in general, and this is not in the book at all, although transformers at the time of writing the book were already in everyone's ears. As a result, the book can be considered an excellent introduction to the basics of AI algorithms, but it should be combined with newer books. When it comes to the content of the book, everything is in it. 10 The chapters I am talking about below

1. The concept of artificial intelligence In this chapter, the author introduces the reader to the basics of artificial intelligence, giving a definition in style that this is a branch of science that is engaged in the development of computer systems capable of performing tasks inherent in human intelligence. Next, he explains the basic concepts, the history of the emergence of AI (term was proposed in 1956 year)and various approaches to the creation of intelligent systems. The author explains the difference between narrow AI (task-maker) general AI (AGI). 2. Basics of search The chapter is devoted to basic search algorithms, which are the foundation for many AI tasks. If you’ve been learning how to code, you can skip this chapter, and if you haven’t, you’ll learn about binary search, breadth and depth search, and how to evaluate the effectiveness of algorithms. 3. Smart search This chapter discusses more advanced search algorithms. The author describes informed search algorithms, such as A search, which uses heuristics to find optimal paths. Algorithms of competitive search are also considered. (min-max search, alpha-beta cut-off)It is used in games and situations where there are opposing parties. These algorithms help the computer make optimal decisions in complex situations. 4. Evolutionary algorithms The chapter introduces evolutionary algorithms, a direction in artificial intelligence that models natural selection processes. Genetic Algorithms That Use Evolutionary Principles (Selection, mutation, interbreeding) To find the best solutions. The author explains how these algorithms work with a population of solutions that “evolve” over time, getting better and better. 5. Advanced evolutionary algorithms This chapter delves into more complex kinds of evolutionary algorithms. Genetic programming is discussed here. (automatic programming)Evolutionary programming (where the program structure is constant and only numerical values change) and other approaches. The author explains the different ways of encoding information in genetic algorithms (Binary, material, ordinal, tree-like) and their application to solve complex optimization problems. 6. Swarm intelligence: ants In this chapter, the author talks about ant algorithms - methods of swarm intelligence that simulate the behavior of ant colonies. It explains how individual ants, being primitive, together form a self-organizing system capable of solving complex problems. We consider how ants use pheromones for communication and how this is applied in algorithms to solve optimization problems, for example, a salesman problem.

Overview of the last four chapters in next post.

#AI #ML #Engineering #PopularScience #Software