Implementation of Evolutionary Algorithms for Deep Architectures

Deep learning is becoming an increasingly interesting and powerful machine learning method with successful applications in many domains, such as natural language processing, image recognition, and hand-written character recognition. Despite of its eminent success, limitations of traditional learning approach may still prevent deep learning from achieving a wide range of realistic learning tasks. Due to the flexibility and proven effectiveness of evolutionary learning techniques, they may therefore play a crucial role towards unleashing the full potential of deep learning in practice. Unfortunately, many researchers with a strong background on evolutionary computation are not fully aware of the stateof-the-art research on deep learning. To close this knowledge gap and to promote the research on evolutionary inspired deep learning techniques, this paper presents a comprehensive review of the latest deep architectures and surveys important evolutionary algorithms that can potentially be explored for training these deep architectures. Index terms — Deep Architectures, Deep Learning, Evolutionary Algorithms