This webinar demonstrated the application of reinforcement learning to create a financial model-free solution to the asset allocation problem, learning to solve the problem using time series and deep neural networks. Using a deep reinforcement model on US stocks and different deep learning architectures, including Long Short-Term Memory networks, Convolutional Neural Networks, and Recurrent Neural Networks. These architectures are compared with more traditional portfolio management approaches like mean-variance, minimum variance, risk parity, and equally weighted. The Deep Reinforcement Learning approach shows better results than traditional approaches using a simple reward function. In Finance, no training to test error generalization results come guaranteed.
Miquel Noguer Alonso Discusses Deep Reinforcement Learning for Asset Allocation in US Equities