Neural Architecture Search

Neural Architecture Search (NAS) is an automated approach for finding the best neural network architecture for a given task. Neural networks are a type of machine learning model that is inspired by the structure and function of the human brain. This chapter presents the Neural Architecture Search (NAS) and its application in cybernetical intelligent systems. It covers various NAS approaches such as Reinforcement Learning‐based NAS, Evolutionary Algorithms‐based NAS, Bayesian Optimization‐based NAS, Gradient‐based NAS, One‐Shot NAS, Meta‐Learning‐based NAS, and NAS for specific domains. The chapter also discusses how NAS can help solve challenges in intelligent control by automating the design and optimization of neural network architectures. Overall, NAS and neural networks are closely related and work together to create more efficient and effective machine learning models. In the future, NAS continue to play an important role in the development of cybernetical intelligent systems.