Neurogenic Deep Learning

Deep neural networks (DNNs) have achieved remarkable success on complex data processing tasks. In contrast to biological neural systems, capable of learning continuously, DNNs have a limited ability to incorporate new information in a trained network. Therefore, methods for continuous learning are potentially highly impactful in enabling the application of DNNs to dynamic data sets. Inspired by adult neurogenesis in the hippocampus, we explore the potential for adding new nodes to layers of artificial neural networks to facilitate their acquisition of novel information while preserving previously trained data representations. Our results demonstrate that neurogenesis is well suited for addressing the stability-plasticity dilemma that has long challenged adaptive machine learning algorithms.