RL-CabDriver | You are hired as a Sr

RL-CabDriver is a Jupyter Notebook library. RL-CabDriver has no bugs, it has no vulnerabilities and it has low support. You can download it from GitHub.

You are hired as a Sr. Machine Learning Er. at SuperCabs, a leading app-based cab provider in a large Indian metro city. In this highly competitive industry, retention of good cab drivers is a crucial business driver, and you believe that a sound RL-based system for assisting cab drivers can potentially retain and attract new cab drivers. Cab drivers, like most people, are incentivised by a healthy growth in income. The goal of your project is to build an RL-based algorithm which can help cab drivers maximise their profits by improving their decision-making process on the field. There are some basic rules governing the ride-allocation system. If the cab is already in use, then the driver won’t get any requests. Otherwise, he may get multiple request(s). He can either decide to take any one of these requests or can go ‘offline’, i.e., not accept any request at all. In this project, you need to create the environment and an RL agent that learns to choose the best request. You need to train your agent using vanilla Deep Q-learning (DQN) only and NOT a double DQN.