Learning and processing framework using Fuzzy Deep Neural Network for trading and portfolio rebalancing

Nicole Hui Lin Kan, Qi Cao, Chai Quek

Abstract: Highlights•An interpretable deep structure IFDNN is proposed to adopt the data driven approach.•The IFDNN is created with fuzzy logic, Hebbian Learning and deep neural networks.•A learning and processing framework is proposed using IFDNN as the trading strategy.•Portfolio rebalancing for financial index assets is conducted with two RL models.