Regime-Augmented Signal Preprocessing for Deep Reinforcement Learning | Quant Club IIT Kharagpur