Neuroevolution (NE) refers to the emerging techniques which combine the search ability of evolutionary computation (EC) and the learning capability of artificial neural networks (ANN) for various tasks, such as finding parameters, hyperparameters and architecture in a typical ANN.

Neuroevolution can be used in all arbitrary neural models and network architectures and in all methods in evolutionary computation to address challenging problems in a sizeable range of domains such as reinforcement learning, supervised learning, unsupervised learning, computer vision, text mining, and speech processing. There are two main methods to combine EC and ANN, first is when we use EC combined with ANN, any parameters, hyper-parameters, and architecture can be optimized by using EC. It not only can use the powerful search ability of EC but also use custom objective function, for instance, the complexity of the model can be minimized, which is hard using conventional optimization algorithms. The second method to combine EC and ANN is when we use ANN combined with EC, the evaluated candidate solutions can be considered as a dataset so that it can estimate the problem model and the fitness landscape, which is essential in computationally expensive problems.

This Special Issue welcomes original research and review articles on the topic of neuroevolution and the combination of EC with ANN.

Potential topics include but are not limited to the following:

• Evolutionary computation algorithms, such as enetic algorithm (GA), particle swarm optimization (PSO), differential evolution (DE), genetic programming (GP), ant colony optimization (ACO) in combination with ANNs
• Neural networks, such as convolutional neural networks (CNN), long short-term memory (LSTM), deep belief network (DBN), feedforward neural networks (FNN), recurrent neural networks (RNN), radial basisfunction neural networks (RBF) in combination with EC
• Novel algorithms for learning the weights of an ANN, and finding the proper hyperparameters
• Evolutionary neural architecture
• Novel search mechanisms
• Surrogate assisted neuroevolution
• Novel representations and objective functions
• Hybrid EC/ANN approaches
• Multi-objective neuroevolution
• Analysis of the complexity of neuroevolution
• Landscape analysis by ANN for EC
• Application of neuroevolution in other scientific fields including image processing and computer vision, text mining and natural language processing, speech processing, software engineering, time series analysis, healthcare, cybersecurity, finance and fraud detection, social networks, recommender systems, and evolutionary robotics