Computational Intelligence (CI) contains the theory, application, and development of computational paradigms motivated by biology and linguistics. Traditionally, there are three main pillars of CI: neural networks, evolutionary computation, and fuzzy systems.
Neural networks, inspired by neural connections, are a subdomain of machine learning and the core of deep learning algorithms, mimicking the way that biological neurons pass signals to one another. There have been a lot of successful applications of neural networks in Artificial Intelligence (AI), such as Computer Vision (CV), Natural Language Processing (NLP), and Reinforcement Learning (RL). Over the past decade, neural-network-based methods have been superior to other machine learning methods that benefited from the increase of computing power and trainable data.
Evolutionary Computation (EC) is a range of biological evolution inspired optimization algorithms. EC techniques can produce highly optimized solutions in a wide range of problem settings. Unlike neural networks, the property of gradient independence makes EC an effective alternative of neural networks in cases where no gradient signal can be utilized. Evolutionary computation is also adopted in some evolutionary biology as an in silico experimental procedure to study common aspects of general evolutionary processes.
Fuzzy system is an approach to calculating "the degrees of truth" rather than the usual "true or false" (1 or 0). In AI systems, fuzzy logic is adopted to mimic human reasoning and cognition. Fuzzy system has been proved as an effective approach in many industrial control fields. It is considered to be a promising method to solve complex comprehension tasks.
Besides the three above main components, CI is also a growing field that includes computational paradigms, such as ambient intelligence, artificial life, cultural learning, artificial endocrine networks, social reasoning, and artificial hormonal networks. CI plays an important role in the development of AI systems which include game and cognitive development systems. In the past few years, research on deep learning prospered, especially on deep convolutional neural networks. Deep learning has become a core method of artificial intelligence.
This Research Topic aims to study the neural network methods inspired by neuro-phenomenon, interaction phenomenon of the biological colony, the evolution of biological populations, etc. Reinforcement learning and self-learning are also the research interests of this topic because they might help reveal the stimulation mechanism of biological behaviors. Submissions that report the latest advances in neural network architectures or methods, neural network applications, and evolutionary computation, will also be encouraged.
Computational Intelligence (CI) contains the theory, application, and development of computational paradigms motivated by biology and linguistics. Traditionally, there are three main pillars of CI: neural networks, evolutionary computation, and fuzzy systems.
Neural networks, inspired by neural connections, are a subdomain of machine learning and the core of deep learning algorithms, mimicking the way that biological neurons pass signals to one another. There have been a lot of successful applications of neural networks in Artificial Intelligence (AI), such as Computer Vision (CV), Natural Language Processing (NLP), and Reinforcement Learning (RL). Over the past decade, neural-network-based methods have been superior to other machine learning methods that benefited from the increase of computing power and trainable data.
Evolutionary Computation (EC) is a range of biological evolution inspired optimization algorithms. EC techniques can produce highly optimized solutions in a wide range of problem settings. Unlike neural networks, the property of gradient independence makes EC an effective alternative of neural networks in cases where no gradient signal can be utilized. Evolutionary computation is also adopted in some evolutionary biology as an in silico experimental procedure to study common aspects of general evolutionary processes.
Fuzzy system is an approach to calculating "the degrees of truth" rather than the usual "true or false" (1 or 0). In AI systems, fuzzy logic is adopted to mimic human reasoning and cognition. Fuzzy system has been proved as an effective approach in many industrial control fields. It is considered to be a promising method to solve complex comprehension tasks.
Besides the three above main components, CI is also a growing field that includes computational paradigms, such as ambient intelligence, artificial life, cultural learning, artificial endocrine networks, social reasoning, and artificial hormonal networks. CI plays an important role in the development of AI systems which include game and cognitive development systems. In the past few years, research on deep learning prospered, especially on deep convolutional neural networks. Deep learning has become a core method of artificial intelligence.
This Research Topic aims to study the neural network methods inspired by neuro-phenomenon, interaction phenomenon of the biological colony, the evolution of biological populations, etc. Reinforcement learning and self-learning are also the research interests of this topic because they might help reveal the stimulation mechanism of biological behaviors. Submissions that report the latest advances in neural network architectures or methods, neural network applications, and evolutionary computation, will also be encouraged.