Master thesis

Portfolio Optimization using Deep Deterministic Policy Gradient and LSTM in Reinforcement Learning

Final report of my master thesis where I engaged in developing a state-of-the-art Deep Reinforcement Learning Method to solve the Portfolio Optimization problem. Studied the impact of neural network architectures (LSTM, CNN, RNN, CRNN) used in DDPG Actor-Critic networks; mixed with alternative data such as financial report, OHLCV or technical indicators.