Deep Q-Network for Stochastic Process Environments

TL;DR

This paper explores the application of Deep Q-Networks in stochastic environments with missing data, using Flappy Bird and stock trading as case studies, and evaluates different network structures.

Contribution

It introduces a systematic evaluation of Deep Q-learning variants tailored for stochastic environments with incomplete information.

Findings

Identified the most effective Deep Q-network structure for stochastic environments

Demonstrated the approach on Flappy Bird and stock trading scenarios

Discussed challenges and proposed future improvements

Abstract

Reinforcement learning is a powerful approach for training an optimal policy to solve complex problems in a given system. This project aims to demonstrate the application of reinforcement learning in stochastic process environments with missing information, using Flappy Bird and a newly developed stock trading environment as case studies. We evaluate various structures of Deep Q-learning networks and identify the most suitable variant for the stochastic process environment. Additionally, we discuss the current challenges and propose potential improvements for further work in environment-building and reinforcement learning techniques.